The field notes library.

Latest Agentforce Naresh Oct 4, 2026 8 min read

Agentforce's new marketing agents: what the Content and Goals agents do

Winter '27 added two marketing agents to Marketing Cloud Next: a Content Agent that drafts on-brand email and SMS, and a Marketing Goals Agent that plans and self-optimizes a whole campaign toward a target you set. One is a drafting tool with a human in the loop. The other asks you to hand over the steering wheel. Here is what each does, where it grounds, what it costs, and the line I would not let the autonomous one cross.

AgentforceMarketing CloudGuardrails

read article →

• 
  AI Agents Naresh Oct 4, 2026 9 min read

  AI agents for SaaS renewals and expansion: automating the motion that moves NRR

  Churn tools tell you an account is dying. By then the renewal is already a fire drill. The higher-value job is the quiet 90 days before: assembling the renewal story, catching expansion signals while they are warm, and freeing the CSM to have the conversation instead of building the deck. Here is how to build a renewal-and-expansion agent on Salesforce, the actions it should own, and the one it must never take alone.

  AI AgentsSaaSCustomer Success
• 
  Data 360 Naresh Oct 4, 2026 10 min read

  Data 360 vs Twilio Segment: the CDP choice when the consumer is an agent

  Both now claim to be the best data layer for AI agents, and they arrive at it from opposite ends. Segment collects events and builds profiles from scratch; Data 360 reads data where it already lives and grounds the agent in the same platform. Here is where each one wins, the pricing models that decide the bill, and the question that settles it.

  Data 360CDPSegment
• 
  Agentforce Naresh Oct 3, 2026 7 min read

  Agentforce A/B experimentation: choosing a winner without rewarding bad answers

  Salesforce announced Agentforce A/B Experimentation for an October beta, with variant setup and an analytics scorecard in Builder. The hard part is deciding what counts as a better agent. Plan an experiment that measures task completion, keeps safety failures visible, and distinguishes a useful change from a different mix of customer requests.

  AgentforceA/B TestingObservability
• 
  Agentforce Naresh Oct 3, 2026 8 min read

  The Agentforce Approval Agent: what it routes, and what it will never decide

  Salesforce built a prebuilt Approval Agent into Agentforce Revenue Management: reps submit and track approvals in conversation, reviewers get AI summaries and approve from Slack. It routes and summarizes. It does not make the call. The real project is the Advanced Approvals engine underneath it, and that is where the gotchas live.

  AgentforceRevenue CloudApprovals
• 
  Data 360 Naresh Oct 3, 2026 9 min read

  Measuring Data 360 retriever quality: how to know your grounding works

  Your agent is grounded and still answers thinly. Before you re-tune chunking again, measure the retriever and the model separately. Here are the two scores that matter, the Data 360 metric object and quality dashboard that now report them, and how to build a golden set that tells a retrieval miss from a generation miss.

  Data 360RAGObservability
• 
  Agentforce Naresh Oct 3, 2026 8 min read

  Prompt Builder conditional logic: testing every branch before activating a template

  Prompt Builder's conditional blocks let grounded data choose which instructions reach the model. That makes one template more flexible and gives you more paths to test. Learn how the beta works, where its collection limit matters, and how to combine branch checks, dependency review, and prompt testing before activation.

  Prompt BuilderConditional LogicTesting
• 
  Platform Naresh Oct 3, 2026 8 min read

  Headless Experience Layer: building AIforce widgets that work across interfaces

  An agent can explain a quote in chat, but a buyer still needs to inspect its lines and confirm the change. Salesforce's Headless Experience Layer brings reusable widgets to Agentforce and external assistants. Learn what the beta supports, how to build the first widget, and which permission and confirmation checks belong in your pilot.

  AIforceHeadless Experience LayerMCP
• 
  Platform Naresh Oct 3, 2026 8 min read

  FORMULA() in the SOQL WHERE clause: comparing two fields at last

  Winter 27 adds a beta FORMULA() function that filters on a computed value inside the WHERE clause, so you stop creating formula fields you never wanted. It is also plus-and-minus only, and sandbox-only. Here is the syntax, the limits, and the one question the docs do not answer.

  ApexSOQLWinter 27
• 
  Platform Naresh Oct 2, 2026 7 min read

  Custom Metadata Types vs Custom Settings: which config belongs where

  You promote a feature-flag change from sandbox to production and every flag comes up off, because the values never deployed. That one behaviour decides most of this choice. Here is what each tool is, the Apex that reads them without spending a query, the write gotcha that keeps Custom Settings alive, and how to pick.

  ApexArchitectureConfiguration
• 
  Platform Naresh Oct 2, 2026 9 min read

  LWC wire service vs imperative Apex: how a component should get its data

  A user saves a contact and the related list two components over still shows the old number. The cause is almost always the wrong data strategy. Here is when to use @wire, when to call Apex imperatively, when to let Lightning Data Service do it, and how to refresh the cache that is lying to your user.

  LWCApexPerformance
• 
  Platform Naresh Oct 2, 2026 9 min read

  Salesforce Platform Cache: Org Cache, Session Cache, and the pattern that makes them safe

  An Apex method rebuilds the same expensive table on every request, for data that changes once a day. Platform Cache fixes that, but it is best-effort storage that can vanish between transactions. Here is Org vs Session Cache, the capacity you get, the CacheBuilder pattern, and the rule that keeps a cache miss from becoming an outage.

  ApexPerformanceArchitecture
• 
  Platform Naresh Oct 1, 2026 9 min read

  Apex tests that assert: coverage is a floor, not a finish line

  A deploy stalls at 74% at 5pm, so someone writes three assertion-free tests to clear the gate, and now the org has coverage that proves nothing. This is how to write Apex tests that verify behavior: build your own data, control limits with startTest, assert real values, mock the outside, and test in bulk.

  ApexTestingCI/CD
• 
  Platform Naresh Oct 1, 2026 11 min read

  Apex trigger frameworks: one trigger per object, and the recursion guard that holds

  Two triggers on the same object fire in an order Salesforce does not promise, and a static boolean recursion flag silently skips half a data load. This is the handler pattern, the Set<Id> guard that survives bulk, the three frameworks worth copying, and the bypass switch your agent writes will need.

  ApexTriggersArchitecture
• 
  Platform Naresh Oct 1, 2026 10 min read

  Making SOQL selective: the non-selective-query error, and the indexes that fix it

  A query that ran fine for a year starts throwing "Non-selective query against large object type" the week your object crosses the threshold. This is what selective really means, the exact index thresholds, the filters that throw your index away, and how to read the query plan before you ship.

  SOQLApexPerformance
• 
  Agentforce Naresh Sep 30, 2026 11 min read

  Agentforce Engagement: the review workflow that decides what the agent sends

  Salesforce renamed Lead Nurturing to Agentforce Engagement in August 2026, but the rename is the small part. The agent now runs on Agent Script, and the human-in-the-loop review criteria you write are what stand between a scheduled outreach and a message a prospect should never have received.

  AgentforceSalesHuman-in-the-Loop
• 
  AI Agents Naresh Sep 30, 2026 10 min read

  AI agents for deal desk: auto-clearing the standard discount, escalating the exception

  A deal desk that runs on a two-day email chain is where fast quotes go to die. An agent can clear the requests that sit inside policy and route the ones that do not, but only if you keep the model out of the one decision that sets a number and protects a margin.

  AI AgentsRevenue CloudSales
• 
  Data 360 Naresh Sep 30, 2026 11 min read

  Connecting Data 360 to Microsoft Fabric: the OneLake bridge, and the beta caveat

  Salesforce shipped connectors for Snowflake, Databricks, BigQuery and Redshift, and the write-ups kept skipping the platform half your Microsoft-shop clients standardized on. Here is the Data 360 to Microsoft Fabric OneLake bridge: federation in versus sharing out, why one side is still beta, and the Entra ID setup that trips Salesforce admins.

  Data 360Zero CopyData Architecture
• 
  Agentforce Naresh Sep 29, 2026 9 min read

  When the customer sends a photo: building image input into an Agentforce agent

  A customer uploads a picture of the cracked device and your text-only agent asks them to describe it. The image was the description. Agentforce can read images and PDFs through a vision-capable model in a prompt template, but the Trust Layer masks text, not pixels. Here's how to wire it, and the governance nobody mentions.

  AgentforceMultimodalEinstein Trust Layer
• 
  Platform Naresh Sep 29, 2026 10 min read

  Grounding your AI coding assistant: the Apex Symbol API in Winter '27

  Your AI assistant writes Apex that looks right and fails to compile, because it pattern-matched on common Salesforce names instead of reading your org's real types. The fix is feeding it compiler-grade metadata. Here's why the hallucination happens, what the SymbolTable already gives you, and what the Winter '27 Apex Symbol API adds.

  ApexAI codingWinter 27
• 
  Platform Naresh Sep 29, 2026 9 min read

  Is Pardot dead? Account Engagement and the path to Marketing Cloud Growth

  No, Salesforce hasn't announced an end date for Account Engagement. It's converging into Marketing Cloud Growth through add-on licenses, not a cutover, and there's no automated migration. Here's what convergence means, what a rebuild costs you, and why most B2B teams should wait.

  Marketing CloudAccount EngagementData 360
• 
  Agentforce Naresh Sep 28, 2026 9 min read

  Agentforce is on by default now: the Winter '27 lockdown checklist

  Salesforce is switching Agentforce on across eligible orgs during the Winter '27 rollout, and removing the toggle that used to keep it off. This is the security pass to run in the week around your upgrade: who can build agents, what they can see, and what is billing.

  AgentforceSecurityWinter 27
• 
  AI Agents Naresh Sep 28, 2026 10 min read

  AI agents for HR service: the confidentiality problem the service playbook ignores

  An HR agent that answers PTO questions is easy. An HR agent that can never tell one employee about another's salary, medical leave, or open investigation is the actual build. Here is the running-user architecture that keeps the boundary, and the topics that must always reach a human.

  AI AgentsHREmployee Service
• 
  Data 360 Naresh Sep 28, 2026 9 min read

  Getting Data 360 fields onto the record: enrichments vs triggered flows

  You unified the data in Data 360, and now a validation rule, a Flow and a report still can't see it, because the field lives in a DMO, not on the Contact. Here are the four ways to bridge that gap, and the rule for picking one.

  Data 360FlowData Architecture
• 
  Agentforce Naresh Sep 27, 2026 10 min read

  Idempotent Agentforce actions: writing the action body so a retry doesn't double-fire

  One customer complaint, three escalation cases. The Atlas reasoning loop can call the same action more than once per intent, and a write action with no guard turns that into duplicate records, duplicate emails, and duplicate charges. This is how to make the action body itself safe to call twice: upsert on an external ID, an idempotency ledger, and the pattern for callouts.

  AgentforceApexActions
• 
  AI Agents Naresh Sep 27, 2026 9 min read

  AI agents for telecom order fallout: fixing the stuck order-to-activate journey

  The broadband order that provisions cleanly never needs an agent. The money is in fallout: the order that stalls between sold and activated, in a system nobody is watching. This is the Communications Cloud object model an agent acts on, the TMF622 change landing in Winter '27, and where the network decision stays with OSS.

  AI AgentsTelecomCommunications Cloud
• 
  Data 360 Naresh Sep 27, 2026 9 min read

  Tuning Data 360 hybrid search: fusion rankers and the knobs that move recall

  Your agent is grounded and still misses the right paragraph. Before you swap the fusion model, fix chunk size and the retriever's result count. This is what Data 360's three fusion rankers actually do, when RRF beats a linear weighting, and the tuning order that moves recall the most.

  Data 360Hybrid searchRAG
• 
  AI Agents Naresh Sep 26, 2026 8 min read

  AI agents for airline revenue accounting: proration, interline settlement, and the order-based shift

  A flown ticket carried by three airlines has to be split across all three, billed through IATA's SIS, and reconciled against a stack of rejection memos. It's rules-heavy, deadline-driven, and mostly manual. Here's where an agent grounded in Data 360 earns its place in the revenue-accounting back office, where it must not, and how the move to order-based settlement changes the target.

  AI AgentsAirlinesData 360
• 
  Platform Naresh Sep 26, 2026 7 min read

  Complex template expressions in LWC: deleting getters you only wrote to format values

  Winter '27 makes complex template expressions generally available in Lightning Web Components. Set apiVersion to 66.0 and a real subset of JavaScript runs in the template, so the getters that existed only to build a CSS class or format a date can go. This is what's allowed, the double-quote rule that trips the compiler, and where a getter still earns its place.

  LWCWinter '27Developers
• 
  Agentforce Naresh Sep 26, 2026 9 min read

  Salesforce Hosted MCP Servers: exposing your Apex and Flows to external agents

  Every MCP post so far treated Salesforce as the client reaching out to someone else's tools. Hosted MCP Servers flip it: your org becomes the server, and Claude, ChatGPT or Cursor call your Apex, Flows and Named Queries under the running user's permissions. Here's the auth model, what you can expose, the global-vs-public trap, and the credit meter behind it.

  AgentforceMCPIntegration
• 
  AI Agents Naresh Sep 25, 2026 9 min read

  AI agents for DSAR fulfillment: automating privacy requests without disclosing the wrong data

  A data subject access request is expensive because finding one person's data across every system is manual, and dangerous because handing it to the wrong person is a breach. An AI agent can carry the search and draft the response. It must never verify identity, disclose, or delete on its own. Here is the split, the Salesforce data model underneath it, and the guardrails that keep an agent from over-disclosing at scale.

  AI AgentsPrivacyCompliance
• 
  Platform Naresh Sep 25, 2026 9 min read

  Salesforce CPQ end of sale: the 2026 migration playbook to Revenue Cloud

  Salesforce CPQ has been end-of-sale since March 2025, frozen but still supported, and the move to Revenue Cloud (Agentforce Revenue Management) is a re-implementation of your pricing logic, not an upgrade. Here is what actually moves, what you rebuild from scratch, and the phased plan that keeps quotes flowing while you do it.

  CPQRevenue CloudMigration
• 
  Platform Naresh Sep 25, 2026 9 min read

  Profiles vs permission sets in 2026: Salesforce cancelled the deadline, not the cleanup

  For three years the plan was to strip permissions off profiles by Spring '26. In June 2026 Salesforce cancelled the retirement in a single edited support article. The forced march is gone, the argument for moving is not. Here is what is still profile-only, how to migrate on your own timeline, and how to avoid trading profile sprawl for permission-set sprawl.

  Permission setsProfilesSecurity
• 
  Agentforce Naresh Sep 24, 2026 10 min read

  Publishing to AgentExchange: what it takes to list an agent action, not an app

  AppExchange is now AgentExchange, and the thing you ship might be a single agent action, not a whole app. Here's how an action, topic, or agent template is packaged as Agentforce metadata, how the security review changes, and what a customer needs licensed to install it.

  AgentExchangeISVAgentforce
• 
  AI Agents Naresh Sep 24, 2026 11 min read

  AI agents for airline refunds: what pays out automatically, and what a human signs

  A DOT refund clock starts the moment a refund is due, and a queue of people can't beat it at scale during a mass disruption. Refunds are deterministic enough for an agent to own end to end. Compensation isn't, because it turns on the extraordinary-circumstances call. Here's the line, the two data points that decide it, and the 2026 rules you're coding against.

  AirlinesTravelCompliance
• 
  Platform Naresh Sep 24, 2026 10 min read

  Einstein Activity Capture: the email store that hides your activity from reports and agents

  Emails that Einstein Activity Capture syncs don't live in your Salesforce database. They sit in a separate store, where your reports, your Flows, Data 360, and your Agentforce agent can't read them. Here's why, what 'Sync Email as Salesforce Activity' changes, and how to get activity data native enough to ground an agent on.

  Einstein Activity CaptureData 360Agentforce
• 
  Data 360 Naresh Sep 23, 2026 10 min read

  Accelerated Data Lake Objects: when zero copy is too slow to ground an agent

  Zero copy federates a query out to Snowflake or BigQuery every time it runs, and your agent inherits that latency. Accelerated Data Lake Objects cache the answer inside Data 360 and refresh it on a schedule. The when-to-accelerate call, when to stay live, the incremental-refresh cost trap, and how data locality decides whether federation is fast or unusable.

  Data 360Zero CopyCredits
• 
  Data 360 Naresh Sep 23, 2026 8 min read

  Document AI in Data 360: turning a PDF into columns an agent can query

  Chunking a contract for retrieval lets an agent read what it says. Document AI does the other job: it pulls named fields, ContractValue, RenewalDate, TotalAmount, out of the PDF into a Data Lake Object your Flows, Calculated Insights, and agents can query. Here is how the schema, the two processing modes, and the licensing work, and where extraction still needs a human check.

  Data 360Document AIAgentforce
• 
  Agentforce Naresh Sep 23, 2026 9 min read

  OAuth token exchange: letting an AI agent act as the user, not the org

  An external AI agent or MCP client needs to do something in your Salesforce org. The lazy answer is a stored integration-user password or a client-credentials token that is the whole org. Both hand a non-human a standing, blanket identity. OAuth 2.0 token exchange is the fix: the agent presents the end user's token and gets back a short-lived Salesforce token scoped to that user. Here is the flow, the Apex handler, and why the audit trail should name a person.

  AgentforceIdentitySecurity
• 
  AI Agents Naresh Sep 22, 2026 10 min read

  AI agents for restaurants and QSR: a guest-service agent on Agentforce and Data 360

  A guest texts 'where's my order' during the dinner rush, and the answer lives in the POS, the loyalty system, and a delivery marketplace you don't own. That gap, not the cooking, is where an agent earns its place. Here is how to build a restaurant and QSR guest-service agent on Agentforce and Data 360, and the two calls that stay human.

  AI AgentsRestaurantsAgentforce
• 
  Data 360 Naresh Sep 22, 2026 10 min read

  Data 360 data retention: expiring Data Lake Objects with no toggle in Setup

  Your data lake grows and nothing in Setup lets you expire the old rows. Retention in Data 360 is a per-DLO lifecycle policy you request through a support case, it only works on Engagement objects, and it keys off the Event Time Field you picked when you built the stream. Get that field wrong and it deletes the wrong records.

  Data 360GovernanceStorage
• 
  Data 360 Naresh Sep 22, 2026 9 min read

  Data 360 webhook data actions: 'Signing Key Not Found' and validating the signature

  Your Data 360 data action is configured, the streaming insight fires, and your endpoint receives nothing. The logs say 'Signing Key Not Found'. Here is why a webhook target stays silent until you generate its secret key, how the HMAC-SHA256 signature works, and the raw-body mistake that makes a correct validation check fail.

  Data 360IntegrationWebhooks
• 
  Agentforce Naresh Sep 21, 2026 10 min read

  When validation rules block your Agentforce agent: the bypass pattern the agent user needs

  The agent worked in the demo and fails on go-live with FIELD_CUSTOM_VALIDATION_EXCEPTION. The cause isn't the model. It's a validation rule written for a human filling a form, firing against a running user that never had the human's context. Here's how to exempt the agent user with a custom permission without weakening the rule for everyone else.

  AgentforceValidation RulesPermissions
• 
  AI Agents Naresh Sep 21, 2026 9 min read

  AI agents for aircraft turnaround: the ground operation an agent can own

  A turnaround is roughly 100 tasks run by about 70 organizations against one clock. The agentic-airport pitch says AI will orchestrate all of it. Here's the honest split: the coordination and exception-handling an agent can own, why A-CDM milestones are the real data model, where Salesforce fits, and the safety line an agent never crosses.

  AI AgentsTravelAgentforce
• 
  Data 360 Naresh Sep 21, 2026 9 min read

  The Data 360 data space tax: why every new brand rebuilds your config

  Stand up a second data space for a new brand or region and you discover the hard truth: none of your DMO mappings, segments, or process definitions come with you. You rebuild them by hand, per space, forever. Data kits are the fix, but the one-click cross-space deployment is sandbox-only. Here's the replication path, what it moves, and what it leaves behind.

  Data 360Data SpacesDevOps
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  Agent Optimizer: the Agentforce tool that reads your traces and proposes the fix

  The most useful demo at Dreamforce 2026 was not a model. It was Agent Optimizer tracing a spike in escalated refunds to a gap in a router, proposing the change, and writing the regression tests. Salesforce dates it for October. This is what it does, what has to be switched on before it can do anything, and how to judge a fix a tool wrote for you.

  AgentforceObservabilityTestingDreamforce 2026
• 
  Agentforce Naresh Sep 20, 2026 7 min read

  Agentforce pricing questions, answered: editions, credits, conversations, and the unpriced parts

  How much does Agentforce cost, what is a Flex Credit, do credits roll over, is Coworker free, and what happens when the pool runs out. Twenty of the pricing questions people search, answered in plain words from Salesforce's published pages as of September 20, 2026, with the parts Salesforce has not priced marked as unpriced.

  PricingFlex CreditsEditionsFAQ
• 
  Agentforce Naresh Sep 20, 2026 9 min read

  Agentforce standard actions: what ships, and when to build your own

  Turn on Agentforce and your agent can already query records, answer from Knowledge, summarise, and draft email before you write a line of Apex. A practitioner read on the standard action library: what each one does, how the reasoning engine picks between them, and the point where a custom action earns its keep.

  AgentforceActionsTopics
• 
  AI Agents Naresh Sep 20, 2026 8 min read

  Agentforce vs. Decagon: the CRM-native agent against the CX-native platform

  Decagon is the best-funded pure-play in AI customer support, and it keeps landing in enterprise bake-offs against Agentforce. A practitioner comparison of where each one is grounded, how Agent Operating Procedures differ from topics and actions, the pricing models, and the question that decides it.

  AgentforceDecagonAI Agents
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  AIforce questions, answered: what it is, what it costs, and what changes for you

  Is AIforce a product, is it free, does it replace Lightning, and which surfaces does it cover today. Twenty of the questions people are searching since Dreamforce 2026, each answered in plain words from Salesforce's own announcement and the coverage around it, with the unpublished parts marked as unpublished.

  AIforceFAQDreamforce 2026
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  AIforce vs Agentforce vs Headless 360: which layer you are buying

  Three names now describe how work reaches Salesforce without a person clicking through Lightning, and they are not three products. Agentforce is agents that act on your data. Headless 360 is the MCP servers and APIs that expose your org to any client. AIforce is those packaged into surfaces you switch on. This is what you configure in each, who the user is, where governance sits, and how each one bills.

  AIforceAgentforceHeadless 360Architecture
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  Amazon Connect and Agentforce Voice: planning for the fall handoff

  One line in the Dreamforce AWS announcement changes the contact centre decision: an agent-to-agent handoff between Amazon Connect and Agentforce Voice, with live audio over WebSockets, dated for fall 2026. This post separates what already works from what is new, lays out the three topologies it makes possible, and lists what to prepare before it ships, without rebuilding your telephony on a promise.

  Agentforce VoiceAmazon ConnectContact centreA2A
• 
  Platform Naresh Sep 20, 2026 8 min read

  Claude Code for Salesforce: the official plugin, the MCP servers, and the skills library

  Salesforce now ships an official plugin for Claude Code. It detects your DX project, loads a set of Salesforce skills, wires in the CLI and Salesforce-hosted MCP servers, and puts safety gates on deploy. This is what it contains, how the three-tier resolution works, what the open skills library is, and the guardrails I would set before letting it near an org.

  Claude CodeDeveloper toolsAgent SkillsMCP
• 
  AI Agents Naresh Sep 20, 2026 7 min read

  Claude Tag vs Slackbot vs an Agentforce agent: which assistant belongs in Slack

  A Slack workspace at a Salesforce customer can now hold three AI members that answer questions and take actions: Anthropic's Claude Tag, Slack's own Slackbot, and an Agentforce employee agent you built. They run on different plans, hold data in different places and act under different permissions. This is the comparison, and the one question that decides which you turn on for which job.

  SlackClaude TagAgentforceAI assistants
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  Commerce Cloud checkout inside Google Search: what UCP changes for merchants in October

  From October, a Commerce Cloud merchant can let a shopper find a product in Google's AI Mode or the Gemini app and buy it there, with the merchant still the merchant of record. The protocol underneath is Google's Universal Commerce Protocol. This is what it does, how it differs from the OpenAI checkout spec, what stays yours, and what to fix in your catalog before the button goes live.

  Commerce CloudAgentic commerceGoogleDreamforce 2026
• 
  Data 360 Naresh Sep 20, 2026 8 min read

  Connecting Data 360 to Databricks with zero copy: federation in, sharing out

  The Data 360 zero-copy bridge to Databricks runs both ways: federate lakehouse tables into Data 360 without moving them, and share unified profiles back into Unity Catalog. The practitioner setup, the File Federation versus Query Federation call, the Unity Catalog and Iceberg prerequisites, and the credit math the "no ETL" pitch leaves out.

  Data 360DatabricksZero Copy
• 
  AI Agents Naresh Sep 20, 2026 7 min read

  What the Dreamforce 2026 customer stories say about production agents

  Salesforce put 21 live customer stories on stage at Dreamforce 2026 and attached numbers to most of them: 96% at the TSA, 95% at Live Nation, 79% at Anthropic, 27,000 employees at Adecco. All of them are the customers' or Salesforce's figures. Read together they show four patterns about what works in production, and what the percentages leave out. This is the reading, plus a checklist for any vendor case study.

  AI AgentsAgentforceCase studiesDreamforce 2026
• 
  Data 360 Naresh Sep 20, 2026 7 min read

  Data 360 at Dreamforce 2026: zero copy everywhere, and the context layer beneath agents

  Data 360 did not get a Dreamforce release. It got pieces of four announcements: zero copy reaching Aurora, RDS, Glue and BigQuery through an Iceberg catalog, Informatica's MCP servers handing agents the golden record, a semantic layer under Tableau, and the session traces that Agent Optimizer will read. This is what a data team can use now, what is dated for the fall, and the inventory to do first.

  Data 360Zero copyInformaticaDreamforce 2026
• 
  Platform Naresh Sep 20, 2026 7 min read

  Dreamforce 2026 for admins: Builder Central, Quick Pages, and the October pile-up

  Strip the keynote down to what changes in Setup and the admin's October gets crowded: Builder Central in beta the week of September 21, Quick Pages in beta in October, Coworker AI Skills and Agent Optimizer going GA the same month, a new Security Center, a compliance agent in Privacy Center, and the Winter '27 production weekends in the middle of all of it. This is the list, with the availability Salesforce has published and the gaps where it has not.

  AdminsBuilder CentralDreamforce 2026Winter '27
• 
  Platform Naresh Sep 20, 2026 8 min read

  Dreamforce 2026 for developers: Claude Code, Agent Script, and what to build first

  The developer story at Dreamforce 2026 was quieter than the keynote and more usable. An official Claude Code plugin backed by the Salesforce skills library, Agent Script as both the way you write an agent and the way Koa was trained, Apex changes that already shipped in Winter '27, and a trace API you can point at your own observability stack. This is the list, what is available today, and what I would build first.

  DevelopersClaude CodeAgent ScriptDreamforce 2026
• 
  Agentforce Naresh Sep 20, 2026 7 min read

  Dreamforce 2026 for sales leaders: Hunter, Piper, Salesforce in Claude, and the rename

  Sales Cloud got its name back at Dreamforce 2026, and three things arrived with it: an outbound agent that works a goal for weeks and is not GA until November, an inbound agent that is GA now, and a Claude plugin with 37 sales skills in beta for every customer. This is what a sales leader can turn on this quarter, what the customer numbers measure, and the order I would do it in.

  Sales CloudAgentforceSalesforce in ClaudeDreamforce 2026
• 
  Agentforce Naresh Sep 20, 2026 7 min read

  Dreamforce 2026 for service leaders: Casey, Fin, the Connect handoff, and what is dated

  Service got the most Dreamforce minutes of any cloud: two named service agents, a rep assistant whose plan rewrites itself, a contact centre product in 30 countries, and a voice handoff with Amazon Connect due this fall. This is what a service leader can use now, what is dated, what the customer numbers do and do not say, and the decisions to make before the October upgrade weekends.

  Service CloudAgentforceContact centreDreamforce 2026
• 
  Platform Naresh Sep 20, 2026 10 min read

  Dreamforce 2026 recap: what shipped, what is dated, and what to do Monday

  Three days, one interface layer, a reasoning model with a name, seven named agents, two hyperscaler deals and a rename reversal. This is the whole of Dreamforce 2026 sorted by what you can use today versus what has a date on it, with a link to a full write-up on each announcement and a Monday list for admins, builders and buyers.

  Dreamforce 2026AgentforceData 360AIforce
• 
  Platform Naresh Sep 20, 2026 9 min read

  The Trusted Enterprise AI Harness: six capabilities, and which product does each today

  Salesforce announced the Trusted Enterprise AI Harness the week before Dreamforce as six capabilities every agent needs, with a unified experience arriving in early fiscal 2028. Most of the six are products you already own under a new diagram. This post maps each capability to what does the job in your org today, what is still a roadmap, and how to use the six as an inventory rather than a purchase.

  Enterprise AI HarnessGovernanceData 360Dreamforce 2026
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  Enterprise Frontier Safeguards: what changes for Claude inside Salesforce

  Anthropic's Enterprise Frontier Safeguards move the activity data used for misuse detection out of Anthropic's systems and into cloud storage you control, under your keys. If your sellers use Salesforce in Claude, your engineers use the Claude Code plugin, or your agents call Claude through Bedrock, this affects you. This is what it is, what it does not change, and the two things to confirm with your account teams.

  ClaudeSecurityZero data retentionClaudeforce
• 
  Agentforce Naresh Sep 20, 2026 7 min read

  Gemini is GA in the Agentforce reasoning engine: how to switch, what to test

  Gemini has been selectable in Agentforce since May and became generally available in the reasoning engine at Dreamforce. Switching is an org-level setting. Deciding whether to switch takes an evaluation set, a look at your grounding, and a view on where your data and your other Google work already live. This is the practical version: what changed, what to run before you flip it, and when Gemini is the right pick.

  GeminiAgentforceModel selectionGoogle Cloud
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  How to test Koa against your current model before the pilot

  Salesforce says Koa makes three times fewer errors on CRM actions. The only number that matters is how it does on your agent's actions, and you can measure that with fifty tasks, the Testing Center and a scorer or two. This is what the published benchmarks measure, what they leave out, and how to build a comparison that turns a pilot into a decision.

  KoaTestingAgentforceModel selection
• 
  Agentforce Naresh Sep 20, 2026 7 min read

  Koa questions, answered: availability, pricing, what it runs on, and who should pilot

  Is Koa available, is it US-only, what is it built on, did Salesforce train it on your data, and what does three times fewer errors mean. Twenty of the questions people are searching about Salesforce's CRM reasoning model, answered from the press release and the research paper, with the unpublished parts left unpublished.

  KoaFAQAgentforceDreamforce 2026
• 
  AI Agents Naresh Sep 20, 2026 7 min read

  Missionforce: Policy Engine, Operations, and Field Operations for government, explained

  The day after the Dreamforce keynote, Salesforce expanded Missionforce, its government platform, with three products: a Policy Engine that turns approved policy documents into deterministic rules, an Operations product for procurement and invoice work in air-gapped environments, and Field Operations with offline mobile. This is what each does, where OpenAI and NVIDIA fit, what is generally available, and the decision that stays with a person.

  MissionforcePublic sectorGovernment CloudDreamforce 2026
• 
  Agentforce Naresh Sep 20, 2026 8 min read

  Salesforce and AWS after Dreamforce 2026: Bedrock models, Amazon Quick, and the Connect handoff

  The Google Cloud announcement got the keynote minutes, but the AWS one changes more for orgs that already run on AWS. Agentforce can now call Anthropic and NVIDIA models through Bedrock, Salesforce data shows up inside Amazon Quick, AWS's own agents arrive in Slack, and a voice handoff between Amazon Connect and Agentforce Voice is due this fall. This is what shipped, what is dated, and what to do with each.

  AWSAgentforceData 360Dreamforce 2026
• 
  Platform Naresh Sep 20, 2026 8 min read

  Slack Code: code channels, the agents inside them, and what security will ask

  Slack Code turns a coding agent into a channel member. Tag Claude, Devin, Copilot or another agent from any conversation and it opens a code channel with diffs, live previews and a searchable record of what it did. This is what ships, which agents are in it, what it costs, and the questions your security team will put to you before the first channel opens.

  SlackCoding agentsDeveloper toolsDreamforce 2026
• 
  Platform Naresh Sep 20, 2026 7 min read

  Slackbot in your channels: Big Mode, voice, Frontline, and what admins control

  Slackbot stopped being a help bot this year. It joins channels as a teammate, has a full-screen mode for long work, gets two-way voice and video generation in the fall, and now has a preview product for shift workers. This is what is live, what is dated, which plans include it, and the settings an admin should check before a million-user feature reaches a workspace.

  SlackSlackbotAI assistantsDreamforce 2026
• 
  Platform Naresh Sep 20, 2026 8 min read

  Winter '27 Flow Builder: Test Mode, multi-record screen flows, and the user-context switch

  Seventeen Flow features land with Winter '27 and three of them change how you work: a test mode that saves and replays tests, screen flows that run against a whole list view, and a run option that enforces the user's permissions no matter who triggered the flow. This is what each does, which are beta, and the two things to do before your production weekend.

  FlowWinter '27AdminsAutomation
• 
  Agentforce Naresh Sep 19, 2026 8 min read

  Authenticating Agentforce actions: Named Credentials, and why per-user fails your agent

  Your agent's action calls an external API and gets a 401. The usual cause isn't the code, it's the authentication model: a per-user credential on an agent that has no interactive user to log in. Here is how outbound auth works for agent actions, and how to set it up so it doesn't break in production.

  AgentforceNamed CredentialsSecurityIntegration
• 
  Platform Naresh Sep 19, 2026 7 min read

  Apex integration tests: testing a real callout instead of a mock of one

  For twenty years an Apex test could not make an HTTP callout, so everyone mocked. Winter '27's @IntegrationTest annotation lets a test hit a real endpoint. Here is what it changes, what it costs you in return, and where mocks still win.

  ApexTestingIntegrationWinter '27
• 
  Data 360 Naresh Sep 19, 2026 9 min read

  Salesforce Connect or Data 360 zero copy: which way to reach external data

  Both let you use data that lives in another system without copying it into Salesforce, and they solve completely different problems. This is the decision: when external objects are the right answer, when Data 360 zero copy is, and when you should just ingest the data after all.

  Salesforce ConnectZero CopyData 360Integration
• 
  AI Agents Naresh Sep 18, 2026 8 min read

  Automating Slack with Agentforce: the actions an agent takes, the workflows that call it

  Most Agentforce-in-Slack advice stops at '@mention the agent and ask it a question.' The automation is the other direction: an agent that posts a canvas or a DM on its own, and a Slack workflow that fires an agent off an emoji or a record change. Here is how both wire together, and the licence and permission traps in between.

  AgentforceSlackAutomation
• 
  Platform Naresh Sep 18, 2026 8 min read

  Salesforce reversed the Agentforce rebrand: your 2026 product-name map

  A year after renaming two dozen products under the Agentforce brand, Salesforce put the old names back on its site the week before Dreamforce, with no announcement. Sales Cloud is Sales Cloud again. Here is the map of what reverted, what kept its new name, and why none of it should change a line of your config.

  Product namingAgentforceEditions
• 
  Agentforce Naresh Sep 18, 2026 9 min read

  Builder Central: Salesforce's new no-code build-and-deploy hub, and where it stops

  The week after Dreamforce 2026, a free build workspace becomes available in every Agentforce-enabled org. You describe an app or an agent in plain language, it reads your metadata, and it deploys to a sandbox in one click. Here is what the beta does, and why it is not your release pipeline yet.

  Builder CentralDevOpsAgentforce
• 
  Agentforce Naresh Sep 17, 2026 9 min read

  AIforce: Salesforce's bet that AI replaces the UI, explained for builders

  At Dreamforce 2026, Salesforce framed AIforce as a live interface layer that pushes your data, workflows, permissions, and governance into Claude, Slack, and your own apps. Here is what it actually is under the branding, when to build against it instead of the Lightning UI, and what changes for security.

  AgentforceAIforceMCP
• 
  Platform Naresh Sep 17, 2026 8 min read

  Salesforce on Google Cloud: what the 'one connected AI stack' changes for architects

  At Dreamforce 2026, Salesforce and Google went from letting their agents call each other to sharing infrastructure: Hyperforce running on Google Cloud, Gemini inside the Atlas reasoning engine, and Salesforce capabilities exposed to Gemini through MCP. Here is what actually shipped, what is still a date on a slide, and how to decide which engine owns a workflow.

  PlatformGoogle CloudAgentforce
• 
  Agentforce Naresh Sep 17, 2026 7 min read

  Slackforce Surfaces: the dashboard a Slack prompt builds, and the permissions it inherits

  Slackforce Surfaces lets anyone describe a dashboard, report, or deck in Slack and have Slackbot build it live from Salesforce data. The build is easy. The governance question is what happens when a surface built under one person's permissions gets posted in a channel of forty. Here is how it works and where to be careful.

  AgentforceSlackSlackforce
• 
  Agentforce Naresh Sep 16, 2026 7 min read

  Salesforce's named agents: when to turn on Casey and Fin, and when to build

  Salesforce gave its agents names and job titles at Dreamforce: Casey for service, Piper for inbound leads, Marshall for supply chain, and four more. Six are generally available, one is in pilot. Here is what a named agent is under the hood, what you can and can't change, and the turn-on-versus-configure-versus-build decision that the branding hides.

  AgentforceAI AgentsPrebuilt agents
• 
  AI Agents Naresh Sep 16, 2026 8 min read

  The AI Control Plane: governing the agents you didn't build and can't see

  Salesforce announced an AI Control Plane at Dreamforce, a layer to register, identify, and govern agents across Salesforce and other platforms, not just your own Agentforce fleet. Here is what it claims to do, how it composes with A2A and MCP, what exists today versus the FY28 roadmap, and the registry discipline worth building now whether or not you buy it.

  AI AgentsGovernanceControl Plane
• 
  Agentforce Naresh Sep 16, 2026 8 min read

  Koa: Salesforce built its own CRM model, and the pitch is fewer wrong actions

  Salesforce announced Koa at Dreamforce, its first CRM reasoning model, post-trained on NVIDIA Nemotron 3 Super. The headline number is not a benchmark leaderboard, it is three times fewer errors on CRM actions. Here is what that claim means, where a purpose-built model beats a bigger general one, and how to slot Koa into the model menu you already have.

  AgentforceLLMKoa
• 
  AI Agents Naresh Sep 15, 2026 10 min read

  AI agents for air cargo booking: quoting, e-AWB, and the ONE Record shift

  Air freight still runs on rate emails, spreadsheets, and a legacy EDI standard IATA started retiring in January 2026. Here's where an agent quotes, books, and documents a shipment across carriers and GSAs, what stays with the carrier's capacity system and CASS, and why ONE Record is the data model that finally makes an agent viable.

  AI AgentsTravelLogistics
• 
  AI Agents Naresh Sep 15, 2026 10 min read

  AI agents for airline crew scheduling: rostering, reserves, and recovery under FAR 117

  Crew is the constraint that turns a two-hour delay into a cancelled bank of flights, and every reassignment has to clear duty-time law before it's legal. Here's where an agent helps a crew controller during IROPS, what stays with the optimizer and the human, and why FAR Part 117 is the boundary the agent proposes up to but never crosses.

  AI AgentsTravelAviation
• 
  Data 360 Naresh Sep 15, 2026 8 min read

  Data 360 in a sandbox: what you can test, and what stays empty

  Data 360 is available in every sandbox now, but the sandbox is metadata-only: your config copies across and your data doesn't, even in Full Copy. Here's what that lets you validate before go-live, what you can't preview without ingesting your own sample, and the provisioning and refresh gotchas that catch teams mid-project.

  Data 360SandboxesDevOps
• 
  Agentforce Naresh Sep 14, 2026 9 min read

  Agentforce Coworker AI Skills: who teaches the agent, and who governs it

  AI Skills let an employee teach Agentforce Coworker a task once and scale it across the workforce. It's the fastest way to capture what your best people know, and the fastest way to scale a mistake. The feature is in pilot with GA planned for October 2026. Here's what an AI Skill is, how it differs from the two other things Salesforce calls a 'skill', and the governance to decide before you turn it on.

  AgentforceCoworkerGovernance
• 
  Agentforce Naresh Sep 14, 2026 10 min read

  Agentforce's long-horizon runtime: building an agent that works a goal for weeks

  Salesforce's new long-horizon runtime lets an Agentforce agent pursue a goal over days and weeks instead of a single chat, and Hunter is the first to run on it. A long-horizon agent isn't a longer conversation. It's a durable plan with checkpoints and an approval boundary, and that changes what you design, test, and watch.

  AgentforceAutonomyArchitecture
• 
  Data 360 Naresh Sep 14, 2026 9 min read

  Connecting Data 360 to Snowflake with zero copy: federation in, sharing out

  Snowflake is where the zero-copy pattern got famous, and it's the connector our BigQuery and Redshift write-ups kept pointing at without walking through. Here's the Snowflake bridge end to end: federation in versus sharing out, the three ways to authenticate now that Salesforce-IdP removes the long-lived key, the warehouse-ownership step that blocks half of first setups, and why Iceberg makes the share-back cleaner than the AWS path.

  Data 360Zero CopySnowflake
• 
  Agentforce Naresh Sep 13, 2026 9 min read

  Adaptive response formats in Agentforce: buttons, carousels, and the v1-only catch

  Your agent knows the answer and delivers it as a paragraph. Adaptive response formats let it render buttons, a carousel, or a rich link instead. The catch that decides your rollout: today they render on the older Enhanced Chat v1 channel, not the new v2.

  AgentforceService AgentMessaging
• 
  Agentforce Naresh Sep 13, 2026 9 min read

  Agentforce in an authenticated Experience Cloud portal: identity, and scoping what the agent sees

  The same agent that's fine on your public website will answer a logged-in member's question with someone else's data if you deploy it the same way. The fix isn't the chat widget. It's who the agent runs as, and which identity you trust when it queries a record.

  AgentforceExperience CloudSecurity
• 
  AI Agents Naresh Sep 13, 2026 8 min read

  AI agents for sales commissions: what Spiff's AI can do, and what it can't

  A rep opens their commission statement, doesn't trust it, and opens the spreadsheet they keep on the side. An AI agent is the right tool for 'why was I paid this' and the wrong tool for computing the payout. That split is the whole design.

  AgentforceSales CloudIncentive Compensation
• 
  AI Agents Naresh Sep 12, 2026 9 min read

  AI agents for meetings and events: what an agent owns, what it can't sign

  The consumer trip-planner agent gets the demo. The harder money is in MICE: venue sourcing, room blocks, and moving 3,000 attendees without a spreadsheet. An agent can run the attendee concierge and draft the RFP. It cannot sign an attrition clause or invent a room rate. Here's the split, the architecture, and where a human still holds the pen.

  AI AgentsTravelAgentforce
• 
  AI Agents Naresh Sep 12, 2026 9 min read

  AI agents for third-party risk: what an agent assembles, what a human accepts

  Third-party involvement in breaches doubled to 30% in a year, and annual vendor reviews can't keep up. An agent can screen a supplier, map its ownership, collect the evidence, and watch the risk feed continuously. It can't accept the risk or terminate the contract. Here's the onboarding-to-monitoring workflow it takes over, and the line it can't cross.

  AI AgentsComplianceFinancial Services
• 
  Data 360 Naresh Sep 12, 2026 9 min read

  Informatica is part of Salesforce now: what it changes for your Data 360 plan

  Salesforce closed the $8B Informatica deal in November 2025, and by May 2026 it was 'Informatica from Salesforce.' The pitch is a trusted data foundation for agents. The practical question is narrower: does it change what you build in Data 360 this quarter? Mostly no, and this is the part that does.

  Data 360MDMGovernance
• 
  Agentforce Naresh Sep 11, 2026 10 min read

  Booking appointments with Agentforce: the Salesforce Scheduler actions, and where the agent stops

  A customer wants to move Tuesday's appointment to Thursday, and most teams reach for Apex to find open slots. If you run Salesforce Scheduler, you don't. Four standard actions read your availability, territories, and work rules, and the whole build comes down to the running user and the channel.

  AgentforceSalesforce SchedulerService
• 
  Platform Naresh Sep 11, 2026 10 min read

  Async Apex: Queueable vs Batch vs future vs Scheduled, and where each breaks

  Four ways to run Apex off the main thread, and most teams pick by habit. The right choice is set by three questions: how many records, do you need to monitor or chain the job, and do you need to keep state. Here's the decision, the limits that bite, and the pattern that keeps an agent action from timing out.

  ApexAsyncGovernor Limits
• 
  AI Agents Naresh Sep 10, 2026 9 min read

  AI agents for utilities: the high-bill call, and the disconnection they can't own

  A cold snap doubles ten thousand bills and the call center drowns in 'why is my bill so high.' An agent grounded on interval usage, weather, and the rate can answer that one well. Disconnection, a reported gas smell, and a hardship deposit are where it has to stop. Here's the split, on Energy & Utilities Cloud and Data 360.

  AI AgentsEnergy & UtilitiesData 360
• 
  Agentforce Naresh Sep 10, 2026 10 min read

  Calling a REST API from Agentforce without Apex: External Services and OpenAPI actions

  Your agent needs to hit Jira, a shipping-quote API, or an internal microservice, and every tutorial reaches for Apex or MuleSoft. If the API has an OpenAPI spec and a Named Credential can log in, you need neither. External Services turns each operation into an agent action with no code.

  AgentforceExternal ServicesIntegration
• 
  AI Agents Naresh Sep 10, 2026 8 min read

  Verifying AI agents across the org boundary: Web Bot Auth, passports, and delegated identity

  Your Agentforce agent-user model answers what an agent can do inside your org. It says nothing about whether an inbound agent is who it claims, or whether your outbound agent can prove who it is. Here's the stack filling that gap, Web Bot Auth, agent passports, verifiable credentials, and delegation tokens, and why it's shipping ahead of the standard.

  AI AgentsIdentitySecurity
• 
  Data 360 Naresh Sep 9, 2026 7 min read

  Consent in Data 360: modeling it, and filtering an activation so the opt-out sticks

  A customer opted out, and the segment still reached them. That's almost never a bad segment. It's consent that was captured but never wired into the activation. Here's how to model consent as a Data Use Purpose and a Contact Point Consent DMO, and how contact point filtering makes the send drop the people who said no.

  Data 360ConsentCompliance
• 
  Agentforce Naresh Sep 9, 2026 11 min read

  The Agentforce token budget: why your agent runs out of room mid-answer

  Every Agentforce turn has a fixed number of tokens to work with, and the model spends them on instructions, history, and grounding before it writes a word. When you overflow that budget the agent doesn't warn you, it drops your safety instructions and answers anyway. Here's what fills the window, how it fails, and how to fit the prompt.

  AgentforceGroundingPerformance
• 
  AI Agents Naresh Sep 9, 2026 9 min read

  AI agents for credit management: clearing the order hold without blowing the credit policy

  A held order is cash you can't recognize and a customer you're annoying, and the release usually waits on a credit analyst assembling exposure by hand. That assembly is an agent's job. The number it lands on, and the limit behind it, are not. Here's the credit-hold workflow an agent can run on Salesforce and your ERP, and the disposition a human still signs.

  AI AgentsFinanceOrder-to-cash
• 
  AI Agents Naresh Sep 9, 2026 10 min read

  AI agents for loyalty fraud: catching the account takeover before the points are gone

  A traveler's miles are cash to a fraudster and a booked liability to you, and the theft is fast, quiet, and mostly aimed at accounts nobody logs into. An agent is well suited to spotting the takeover in real time and badly suited to deciding the clawback. Here's the detection-and-triage workflow on Salesforce Loyalty Management and Data 360, and the reversal a human still signs.

  AI AgentsTravelFraud
• 
  Agentforce Naresh Sep 8, 2026 8 min read

  Grounding an Agentforce agent on data you can't ingest: External Objects and live calls

  Your agent has to answer from data that lives in another system and changes by the minute: live stock, an order's status, today's price. The grounding tutorials assume that data is already in Data 360. Here's the decision they skip: when to ingest, when to virtualize with External Objects, and when to just call the API live, plus the one constraint that rules most of it out.

  AgentforceData 360Integration
• 
  Agentforce Naresh Sep 8, 2026 8 min read

  Reward hacking in Agentforce: when your agent games the metric

  Give an agent a single number to maximize and it will find the cheapest path to that number, which is rarely the outcome you meant. An agent measured on cases closed learns to close cases the moment a customer goes quiet. This is reward hacking, and no guardrail catches it. Here's why the Trust Layer won't stop it, and how to design metrics, actions, and tests an agent can't game.

  AgentforceGuardrailsAgent testing
• 
  AI Agents Naresh Sep 8, 2026 8 min read

  AI agents for short-term rentals: the one to build, and the one to buy

  Short-term rental managers already have AI agents inside Guesty and Hostaway. So the real question for a large operator is not whether to run an agent, but whether to build one on Agentforce and Data 360 or buy the one that ships with the PMS. Here's the split that decides it, the data foundation underneath, and the OTA rules that break a generic agent.

  AI AgentsHospitalityAgentforce
• 
  Data 360 Naresh Sep 7, 2026 10 min read

  Where your Agentforce conversation transcripts live, and how to get them out

  The message your agent sent a customer may not be in Salesforce at all. Enhanced-channel and messaging transcripts live in a separate off-platform store you cannot reach with SOQL, agent reasoning traces expire in 72 hours, and the Trust Layer audit log only exists if you turned it on. This is the map of where conversation data goes, the three ways to get it out, and what deletion has to reach.

  Data 360AgentforceCompliance
• 
  AI Agents Naresh Sep 7, 2026 11 min read

  AI agents for passenger rail: assistance bookings, delay claims, and the safety line

  Rail is not aviation with tracks. The high-volume, rule-bound work an agent can take on is booking passenger assistance, answering disruption questions from live running data, and starting a Delay Repay claim. This is each workflow, the data it has to ground on, and the duty-of-care line a human still confirms.

  AI AgentsTravelAgentforce
• 
  Agentforce Naresh Sep 7, 2026 10 min read

  Claudeforce questions, answered: pricing, availability, models, and what changes for you

  Is Claudeforce free, when can you get it, does it replace Agentforce, which Salesforce products now run on Claude, and where does your data go. Twenty of the questions people are searching about the Salesforce and Anthropic partnership, each answered in plain words from the published sources, with the parts nobody has published marked as unknown.

  ClaudeforceFAQAgentforce
• 
  Agentforce Naresh Sep 7, 2026 10 min read

  The new Salesforce editions: what Core, Advanced and Max change for your budget

  On September 3, 2026, Salesforce collapsed its edition sprawl into three tiers, Core, Advanced and Max, and bundled Agentforce, Slack, Tableau Next and a pool of Flex Credits into each. This is what each tier costs, what it replaces, what is tier-gated, and the credit arithmetic the bundle makes easy to under-budget.

  AgentforcePricingFlex Credits
• 
  Agentforce Naresh Sep 6, 2026 11 min read

  Agentforce vs. Lyzr: the CRM-native agent layer against the deploy-anywhere platform

  Lyzr is the Agentforce alternative regulated enterprises keep shortlisting: full-stack agent infrastructure that runs in your own VPC, routes to any model, and prices per run. Agentforce is the agent layer of the CRM your sellers and service reps already live in. This is the comparison across hosting, grounding, model freedom, build model, governance and pricing, and the one question that decides it.

  AgentforceLyzrComparison
• 
  Agentforce Naresh Sep 6, 2026 10 min read

  Salesforce in Claude: what the 37 skills do, and how to pilot the beta

  The Claudeforce plugin that puts a governed Salesforce org inside a Claude conversation goes to open beta this month, with a live demo at Dreamforce. This is the practitioner's guide: the five jobs the 37 skills cover, the hosted MCP server and per-user OAuth underneath, what a write from Claude triggers in your org, the zero-data-retention story, and a four-week pilot that ends in a decision.

  ClaudeforceMCPAgentforce Sales
• 
  Platform Naresh Sep 6, 2026 11 min read

  Salesforce Winter '27 for Agentforce and Data 360 builders: what changes and what breaks

  Winter '27 is already live on the first production instances and Dreamforce lands between the second and third upgrade weekends. This is the practitioner's cut: the Agentforce, Data 360 and Apex changes that alter how you build, the release updates and API retirements that will break an integration, and what to test in the preview sandbox before your org's date.

  Winter '27Release notesAgentforceData 360
• 
  Agentforce Naresh Sep 5, 2026 9 min read

  Agentforce DX: building and previewing agents from the CLI instead of clicking through Studio

  Every agent you build in the browser is a config you cannot diff, review, or roll back. Agentforce DX turns an agent into source: a YAML spec, generate and preview commands, and metadata that moves through your pipeline like any other. Here is the pro-code loop, the metadata under the hood, and the deployment dependencies that trip up the first migration.

  AgentforceDevOpsDeveloper tools
• 
  AI Agents Naresh Sep 5, 2026 9 min read

  Payments are the hard part of an AI travel agent: PCI scope, network tokens, and the two-sided settlement problem

  The conversation is the easy 20%. The moment your travel agent has to take a card and pay a supplier, you inherit PCI-DSS scope, a two-sided settlement problem the retail agentic-commerce playbook never mentions, and a stack of brand-new agent-payment protocols. Here is how an agent takes money without ever touching a card number, and where the airline and hotel legs still break.

  AI AgentsTravelPayments
• 
  Platform Naresh Sep 5, 2026 10 min read

  Apex Cursors: processing millions of records without Batch Apex, and the limits that still bite

  Spring ’26 made Apex Cursors generally available, and the reflexive reaction is "Batch Apex is dead." It is not, but there is now a better tool for a large class of high-volume jobs. Here is what a cursor is, the Queueable pattern that makes it work, the four governor limits nobody reads until they hit one, and when to still reach for Batch.

  ApexPerformanceGovernor limits
• 
  Agentforce Naresh Sep 4, 2026 14 min read

  Agentforce dynamic plans: when the resolution rewrites itself mid-case

  Winter '27's Adaptive Experiences point an ambient agent at a live case, and it regenerates the resolution plan as the conversation changes, no new prompt required. Here's what a 'dynamic plan' is, why it's a rep-assist and not an autonomous agent, and the latency, credit, and plan-thrash trade-offs you inherit the moment the plan stops holding still.

  AgentforceDynamic PlansService Cloud
• 
  AI Agents Naresh Sep 4, 2026 13 min read

  Getting your hotel bookable by an AI agent: the agent-ready distribution stack

  Google now books hotels inside AI Mode, ChatGPT plans trips with live pricing, and the launch partner list was OTAs and the big brands, not the independents. The question has flipped from 'should we build an agent' to 'can an agent buy from us.' Here's what agent-ready distribution requires: real-time ARI, machine-readable rates, delegated payment, and where a unified guest profile is the difference between getting booked and getting bypassed.

  AI AgentsTravelHospitality
• 
  Data 360 Naresh Sep 4, 2026 14 min read

  The Data 360 credit circuit breaker: the kill-switch Digital Wallet won't give you

  Digital Wallet will email you at 90% of your credit threshold. It will not stop the runaway job that's burning them. Here's how to build the missing half: a Flow-plus-Apex breaker that reads live consumption and clears the schedule on the offending workload before it bills, and the four gotchas that make an automated pause trickier than it looks.

  Data 360CreditsFlow
• 
  Agentforce Naresh Sep 4, 2026 13 min read

  Why your Agentforce agent feels slow, and how to get the first token back faster

  A pilot that dazzles on a clean demo turns sluggish in production, and everyone blames the model. The real cost is usually the reasoning loop, the grounding, and the actions, not the LLM. Here is what happens between a customer pressing enter and the first word appearing, where the seconds go, and the levers that pull them back.

  AgentforcePerformanceArchitecture
• 
  AI Agents Naresh Sep 4, 2026 12 min read

  AI agents for car rental: the agent is easy, the fleet and the folio are the build

  A rental agent that quotes the cancellation policy is a demo. One that checks live fleet availability, extends a rental, dispatches roadside assistance, and reads the charge a customer is disputing is a product, and the hard part is never the model. It's the reservation system, the telematics feed, and the read-versus-write line that decides whether you shipped an asset or a liability.

  AI AgentsTravelAgentforce
• 
  Data 360 Naresh Sep 4, 2026 12 min read

  Data streams in Data 360: the category choice you cannot undo, and the refresh mode that silently drops deletes

  Getting data into Data 360 looks like a wizard: pick a connector, map some fields, done. Two of those clicks are load-bearing in a way the UI never warns you about: the stream category is permanent, and the refresh mode decides whether deletions in the source ever reach your unified profile. Here is the connector landscape, the two decisions that are hard to reverse, and when not to ingest at all.

  Data 360IngestionArchitecture
• 
  Agentforce Naresh Sep 3, 2026 11 min read

  AI agents that write your knowledge base: Einstein Knowledge Creation and the closed loop with Agentforce

  Everyone grounds their Agentforce agent on the knowledge base. Almost nobody feeds it. Einstein Knowledge Creation drafts an article from a resolved conversation: reviewed by a human, never auto-published, so the knowledge that answers the next customer is captured as a by-product of solving this one. Here's the setup, the closed loop into Agentforce, and the governance that keeps a bad draft from poisoning the agent.

  AgentforceKnowledgeService Cloud
• 
  AI Agents Naresh Sep 3, 2026 11 min read

  AI agents for car dealerships: what Agentforce Automotive can run, and the calls the dealer still makes

  Agentforce for Automotive runs on Automotive Cloud's vehicle-and-warranty data model and grounds on telematics unified in Data 360. It can book the service, draft the quote, and coordinate inventory across stores. What it can't do is quote a final price, adjudicate a warranty, or promise a loan approval, because the FTC treats every claim it makes as the dealership's. Here's the architecture, and the line.

  AI AgentsAutomotiveAgentforce
• 
  Agentforce Naresh Sep 3, 2026 11 min read

  Salesforce is buying Fin (formerly Intercom): what the $3.6B deal means for your Agentforce roadmap

  On June 15 2026 Salesforce signed a definitive agreement to acquire Fin, the company that was Intercom until May, for about $3.6 billion. It's the clearest signal yet about where the service-agent market is heading, and it changes the build-vs-buy math for anyone evaluating an AI support agent right now. Here's what's real, what's still just signed-not-closed, and how to decide today.

  AgentforceFinService Cloud
• 
  Agentforce Naresh Sep 2, 2026 13 min read

  The Agentforce inbound lead-generation agent: turning anonymous website visitors into booked meetings

  Salesforce shipped a pre-built agent that greets the stranger on your pricing page, answers their questions from your own content, qualifies them into a Lead, and books time on a rep's calendar, all before a human sees it. Here's how the Lead Generation agent works, how it differs from the SDR agent everyone confuses it with, and the injection risk you inherit the moment you point an agent at anonymous input.

  AgentforceSalesLead generation
• 
  AI Agents Naresh Sep 2, 2026 13 min read

  AI agents for hotel revenue management: an RM copilot on Agentforce and Data 360

  Your RMS already does the pricing math. What eats the revenue manager's morning is everything around it: assembling the pickup report, shopping the comp set, explaining what moved and why. That's the part an agent is good at. Here's how to build an RM copilot on Agentforce and Data 360 that recommends and drafts, with a human on the rate button.

  AI AgentsHospitalityRevenue management
• 
  Data 360 Naresh Sep 2, 2026 11 min read

  Data 360 vs MDM: does the unified profile replace your golden record?

  The most expensive assumption in a Data 360 rollout is 'we bought this, so we can retire MDM.' Data 360 links records into a unified profile for activation; MDM masters them into a governed golden record you can defend to an auditor. Here's the real difference, where each stops, and the pattern where both earn their keep.

  Data 360MDMArchitecture
• 
  AI Agents Naresh Sep 1, 2026 12 min read

  AI agents for accounts receivable: collecting cash without breaking the ledger

  Accounts payable gets the agent headlines, but the money problem is on the other side of the ledger, cash stuck in receivables while someone chases it by hand. Here is the AR workflow an agent can own, the cash-application matching that decides whether it works, and the postings a human still has to sign.

  AI AgentsFinanceOrder to Cash
• 
  AI Agents Naresh Sep 1, 2026 11 min read

  AI agents and airline retailing: selling offers and orders an agent can trust

  The airline industry is tearing out the fare, the ticket, and the PNR and replacing them with offers and orders, right as AI shopping agents arrive to buy them. Here is what NDC and ONE Order change, where an agent fits in an airline retailing stack, and the hard line where the offer engine owns the price and Salesforce does not.

  AI AgentsTravelRetailing
• 
  AI Agents Naresh Sep 1, 2026 10 min read

  AI agents for customer onboarding: closing the time-to-value gap without faking activation

  Most B2B onboarding is a checklist a human chases and a customer ignores, and half of signups never reach value. An agent can drive onboarding tasks around the clock, but "activated" is a real outcome you cannot fake by ticking a box. Here is the architecture that shortens time-to-value, the signals it runs on, and the escalation gate that keeps a human in the loop.

  AI AgentsCustomer SuccessOnboarding
• 
  Agentforce Naresh Aug 30, 2026 9 min read

  Agentforce on mobile: the built-in agent, the Mobile SDK, and the offline wall nobody warns you about

  There are two completely different ways to put an Agentforce agent in someone's hand, and teams keep conflating them. One ships with zero build; the other is an SDK you embed in your own app. Here's how each works, the auth model under the SDK, and the connectivity limit that decides where a mobile agent is worth building at all.

  AgentforceMobileSDK
• 
  Agentforce Naresh Aug 30, 2026 10 min read

  Agentforce for Public Sector: what building agents in Government Cloud changes

  FedRAMP High authorization put autonomous agents on the table for government agencies, but Government Cloud is not commercial Salesforce with a compliance sticker. The model menu is restricted, features arrive late, and the one decision an agent can never make is the one that determines a citizen's benefits. Here's the practitioner's map.

  AgentforcePublic SectorCompliance
• 
  Agentforce Naresh Aug 30, 2026 11 min read

  Claudeforce, explained for practitioners: what changed when Anthropic moved inside the Salesforce trust boundary

  Salesforce and Anthropic announced Claudeforce on August 26, 2026. Behind the branding are three concrete architectural changes: Claude runs inside the Salesforce VPC, it becomes the default reasoning model for several agent products, and Salesforce ships as a plugin inside Claude. Here's what each one means for how you build.

  AgentforceLLMMCP
• 
  Agentforce Naresh Aug 30, 2026 12 min read

  Agentforce limits that shape your architecture: agents, subagents, actions, and the governor-limit collision

  Nobody hands you the caps up front, 20 active agents per org, 15 subagents per agent, 15 actions per subagent, a 120-second API turn, a six-turn memory window. Here's the full limit map, why the Atlas reasoning loop burns Apex governor limits differently than a trigger does, and the async patterns that survive it.

  AgentforceArchitectureLimits
• 
  AI Agents Naresh Aug 30, 2026 12 min read

  AI agents for corporate travel duty of care: locating every traveler when a crisis hits, and the alert the agent never sends alone

  When an earthquake, a coup, or a grounded fleet hits a city your people are in, the question is brutally simple: who's there, and are they safe? A triggered Agentforce agent can watch the risk feed, match it to live itineraries, and draft the check-in, but the reassurance that says 'stay put' is the one call it must never make alone.

  AI AgentsTravelAgentforce
• 
  Data 360 Naresh Aug 30, 2026 11 min read

  Problem Records in Data 360: why records vanish on ingestion, and how to get them back

  Your data stream shows green, the run 'succeeded,' and the row count is lower than your source. The missing rows didn't error the batch: they got quarantined into a PR_ Problem Records object you didn't know to look at. Here's the error taxonomy that puts them there, how to diagnose it, and how to reprocess without duplicating.

  Data 360IngestionTroubleshooting
• 
  Agentforce Naresh Aug 29, 2026 10 min read

  Agentforce vs. Sierra: the CRM-native agent vs. the brand-owned CX layer

  Sierra is the most-searched Agentforce alternative that isn't another hyperscaler, a standalone customer-experience agent platform from Bret Taylor, sold on outcomes, wired to whatever stack you have. Agentforce is the agent layer of your CRM. They win in different worlds. Here's the honest split on architecture, grounding, actions, pricing, and the one question that settles most of these evaluations.

  AgentforceSierraComparison
• 
  AI Agents Naresh Aug 29, 2026 10 min read

  AI agents for cruise lines: the onboard-revenue agent, and the calls it never gets to make at sea

  A cruise isn't a long hotel stay, it's a closed, multi-day environment where more than half of onboard spend is booked before boarding, connectivity drops the moment you leave port, and the captain, not the agent, decides when the itinerary changes. Here's the Salesforce architecture for a guest agent that plans, upsells, and re-accommodates a missed port, and the safety and policy boundaries it must never cross.

  AI AgentsTravelAgentforce
• 
  Data 360 Naresh Aug 29, 2026 11 min read

  Data freshness in Data 360: why your unified profile is older than you think, and how to monitor the whole chain

  A data stream that says 'refreshed 2 minutes ago' can still be grounding your agent on yesterday's truth, because freshness isn't one number, it's the slowest link in a chain of five stages that each run on their own clock. Here's how staleness hides in Data 360, the silent failures that never throw an error, and a practical way to monitor freshness end to end before an agent quotes a stale number to a customer.

  Data 360Data qualityGovernance
• 
  Agentforce Naresh Aug 28, 2026 13 min read

  Agentforce Coworker: the AI teammate Salesforce is turning on in your search bar

  Salesforce started auto-enabling Agentforce Coworker in Winter '27, a conversational layer that answers from your data and routes the next step to an agent. Here is what it is, how it differs from an Assistant and an Employee Agent, the permission-set gotcha that decides whether it burns credits, and how to govern a feature that switches itself on.

  AgentforceCoworkerLicensing
• 
  AI Agents Naresh Aug 28, 2026 12 min read

  AI agents for mishandled baggage: the claim an agent can own, and the liability call it cannot

  A delayed-bag claim is mostly status lookups, receipt checks, and reassurance, work an agent handles well. But "declare this bag lost" and "settle this liability" are decisions with a treaty behind them. Here is how to build a baggage agent on Salesforce that owns the 90% and knows exactly where the WorldTracer status ends and a human judgment begins.

  AI AgentsTravelService Cloud
• 
  Platform Naresh Aug 28, 2026 12 min read

  Salesforce Data Detect: finding the sensitive data hiding in your text fields before an agent reads it

  You cannot govern data you have not found. Data Detect: the fourth pillar of Shield, scans free-text fields across your org for PII using prebuilt and custom patterns. Here is exactly which fields it reads, the 200-row wall in the UI, how to pull ten million results with SOQL, and the honest limitation nobody mentions: it classifies, it does not clean up.

  SecurityShieldData 360
• 
  AI Agents Naresh Aug 27, 2026 13 min read

  AI agents for tour operators: dynamic packaging when no single supplier owns the trip

  A flight, a hotel, a transfer, and an activity assembled into one trip is a distributed-transaction problem no single API solves, and an agent that books a rate which already moved is a liability, not a feature. Here's the multi-agent architecture that packages across suppliers, revalidates before it commits, and where a human still has to sign.

  AI AgentsTravelAgentforce
• 
  Agentforce Naresh Aug 26, 2026 11 min read

  Agentforce vs. OpenAI: a stack-honest comparison for building enterprise AI agents

  One is the agent layer of a CRM; the other is the agent layer of the model frontier. They're both real ways to build enterprise agents, and they win in different worlds. Here's where each fits: grounding, governance, build model, cost, lock-in, and the tiebreaker question that settles most evaluations.

  AgentforceOpenAIComparison
• 
  Data 360 Naresh Aug 26, 2026 11 min read

  Agentic setup in Data 360: what an agent can scaffold in your pipeline, and the calls it can't make

  Salesforce is bringing natural-language, agent-assisted setup to Data 360, describe the pipeline and it proposes streams, transforms, mappings, and segments. That collapses the busywork. But the handful of decisions that determine whether your data is right are exactly the ones an agent optimizes into a plausible, wrong configuration. Here's what to hand it, and what to keep.

  Data 360AgentforceIdentity Resolution
• 
  Agentforce Naresh Aug 26, 2026 11 min read

  Agentforce Custom Scorers: grading an agent session against your own definition of resolved, not Salesforce's

  The built-in metrics tell you whether a response was helpful and whether the customer escalated. They can't tell you whether this specific session accomplished the outcome your use case exists to produce. Custom Scorers: new in Winter '27, let you write that pass/fail question as a prompt-template scorer and run it against real production sessions. Here's the AiAgentScorerDefinition metadata, the engine-and-output-mapping shape, and where it fits next to the CI tests you already run.

  AgentforceObservabilityTesting
• 
  AI Agents Naresh Aug 26, 2026 12 min read

  AI agents for travel document and visa compliance: the pre-departure readiness check that never overrules the border

  A passport three months from expiry, a missed transit-visa rule, a UK ETA nobody applied for, document gaps get caught at the gate, where they're most expensive. An Agentforce agent grounded on an authoritative entry-requirements feed can flag them days earlier. Here's the architecture: Timatic or Sherpa as the source of truth, Data 360 for the traveler profile, and the guardrail that keeps the agent advisory, because the airline and the border make the real call.

  AI AgentsTravelAgentforce
• 
  Data 360 Naresh Aug 25, 2026 12 min read

  Querying Data 360 from Apex: the new sfsqlquery namespace, and the SQL-vs-SOQL line that trips everyone

  Winter '27 adds a first-class way to run Data 360 SQL from Apex, the sfsqlquery namespace, with synchronous, resumable, and Queueable-paged workflows that replace hand-rolled ConnectApi.CdpQuery loops. Here's what the new classes do, the ConnectApi path they supersede, why it's SQL on the Hyper engine and not SOQL, and the governor limits and credits that decide how you write the query.

  Data 360ApexData 360 SQL
• 
  Agentforce Naresh Aug 25, 2026 13 min read

  Running an Agentforce agent in more than one language: the allowlist, Translation Workbench, and the grounding gap that answers in English anyway

  The model already speaks forty languages, so teams assume multilingual is a toggle. It isn't. 'Support Spanish' is three separate problems: a language allowlist, a language signal, and grounding that filters your knowledge out when it's in the wrong language. Here's what carries a conversation in a second language, and the one gap that makes an agent answer in English no matter what you configured.

  AgentforceMultilingualLocalization
• 
  AI Agents Naresh Aug 25, 2026 14 min read

  AI agents for higher education: the student-services agent on Education Cloud, and the decisions it never gets to make

  A student at 11 p.m. the night before the registration deadline can't reach anyone, and there's a hold on their account they don't understand. That gap: 24/7 demand meeting a shrinking, overworked staff, is where a student-services agent earns its keep. Here's the Education Cloud data model it runs on, the enrollment-to-advising work it can own, and the regulated line where financial aid, admissions, and grades stay human.

  AI AgentsHigher EducationAgentforce
• 
  AI Agents Naresh Aug 25, 2026 15 min read

  AI agents for travel insurance claims: the flight-delay payout the agent can own, and the adjudication it can't

  A flight delayed three hours can pay its own claim before the passenger lands, no form, no adjuster. A lost-bag claim can't, and shouldn't. That split, parametric versus indemnity, is the whole design. Here's the travel-claims work an agent owns, the PIR-and-refund-offset detail a generic claims agent misses, and the licensed-adjuster line the agent never crosses.

  AI AgentsTravelInsurance
• 
  Agentforce Naresh Aug 23, 2026 11 min read

  Building a guided shopping agent in Agentforce Commerce: the Shopper, Buyer, and Merchant agents, and the SCAPI plumbing under the storefront

  Agentforce Commerce went GA in June 2026 with three agents, Shopper, Buyer, and Merchant. But the demo where an agent recommends a jacket hides the real build: the SCAPI and SLAS authorization that lets it read a live cart, the topics that keep it from inventing a promo code, and the difference between the managed agents Salesforce ships and the guided shopping agent you configure. Here's the architecture.

  AgentforceCommerceData 360
• 
  AI Agents Naresh Aug 23, 2026 11 min read

  AI agents for warranty claims: adjudicating against the entitlement without rubber-stamping the fraud

  A warranty claim isn't an insurance loss or a return, it's an entitlement decision: is this asset, this failure, this cost covered? Manufacturing Cloud ships the object model to answer that, and Agentforce can create the claim, validate it, and route the exceptions. Here's where an agent safely clears the routine claims, and the line: high value, suspected fraud, goodwill, it must never cross alone.

  AI AgentsManufacturingAgentforce
• 
  Data 360 Naresh Aug 23, 2026 11 min read

  Connecting Data 360 to Amazon Redshift: the zero-copy connector, IdP auth, and the two directions data can flow

  You already have the Snowflake and BigQuery bridges. Redshift is the one people skip, and its setup is different: OpenID Connect trust, an IAM role with an external ID, and a JDBC URL, not a service account. Here's the federation connector step by step, the two directions zero copy runs, and the gotchas that decide whether a live query is fast or a governor-limit disaster.

  Data 360Zero CopyAWS
• 
  Agentforce Naresh Aug 23, 2026 10 min read

  Agentforce Help Agent and pay-per-resolution: what changes when you only pay for outcomes

  Salesforce shipped Agentforce Help Agent in June 2026 with a pricing model that breaks from Flex Credits entirely: $2 per resolution, nothing when the agent fails or escalates. It's the biggest change to Agentforce economics since launch, and 'you only pay when it works' is a headline, not a contract. Here's what a resolution is, the accuracy gap nobody markets, and how to model the cost honestly.

  AgentforcePricingService Cloud
• 
  Agentforce Naresh Aug 23, 2026 12 min read

  Agentforce Operations: what changes when the agent works the back office, not the case queue

  Agentforce spent a year automating the front office, service, sales, marketing. Agentforce Operations, launched in April 2026, points the same agents at finance, procurement, and supply chain, and it introduces a 'process blueprint' as the thing agents run against. Here's what it is, how it differs from a Flow, where the ERP-integration reality bites, and how to think about consumption pricing before you turn it on.

  AgentforceAutomationEnterprise AI
• 
  Data 360 Naresh Aug 23, 2026 12 min read

  Open Semantic Interchange: the standard that stops your Data 360 metrics from being trapped in one tool

  You defined 'net revenue' once in Tableau Semantics, again in dbt, and a third time in a Snowflake view, and an agent grounded on any of them gives a different number. Open Semantic Interchange, now Apache Ossie, is the vendor-neutral YAML standard built to end that. Here's what it standardizes, what it deliberately doesn't, and what it means for a Salesforce data team.

  Data 360Semantic layerData architecture
• 
  AI Agents Naresh Aug 21, 2026 11 min read

  AI agents for complaint management: where the agent triages and drafts, and where a regulated response stays human

  A regulated complaint isn't a support ticket, it starts a statutory clock, and mis-tagging one as a general query is a compliance failure, not a CX miss. An agent is good at the intake, evidence-gathering, drafting, and the thematic root-cause analysis most firms underuse. Here's the workflow it can own on a Salesforce stack, the regulatory lines it can't cross, and why classification is the highest-stakes step.

  AI AgentsFinancial ServicesCompliance
• 
  Platform Naresh Aug 21, 2026 12 min read

  Salesforce backup and restore: why the Recycle Bin isn't a backup, and what native Backup protects

  The Weekly Export and the Recycle Bin feel like a safety net until the day a bad deployment or a rogue mass-update wipes real data, and you learn the export has no relationships, the Recycle Bin emptied itself two weeks ago, and Salesforce retired its $10k last-resort recovery years ago. Here's the shared-responsibility reality, what Salesforce's native backup covers, and why the restore, not the backup: is the part that goes wrong.

  Backup & RestoreData ProtectionDisaster Recovery
• 
  Platform Naresh Aug 21, 2026 11 min read

  Salesforce Hyperforce: what the migration changes, and the integrations it breaks

  Since July 1 2026 you can no longer delay the move to Hyperforce, Salesforce schedules it into your maintenance window whether you're ready or not. The platform gets faster and lands in more countries, but the hardcoded instance URL, the IP allowlist, and the pinned certificate that worked for a decade are exactly what break on cutover. Here's what Hyperforce is, what the migration does to your org, and the prep that decides whether go-live is boring.

  HyperforceData ResidencyMigration
• 
  Luxury Travel Naresh Aug 18, 2026 17 min read

  Virtuoso Travel Week 2026: the full recap of what happened in Las Vegas

  The 38th Virtuoso Travel Week drew 4,800 attendees from 105 countries to Las Vegas for 110,000+ one-to-one meetings. Here's the complete recap: the sales numbers, the 'fall is the new summer' data, the AI announcement, the stewardship launch, and every award winner.

  VirtuosoLuxury TravelTravel Trade
• 
  Agentforce Naresh Aug 17, 2026 14 min read

  The Agentforce agent user: which user your agent runs as, and why it can't update the record

  The most common Agentforce failure isn't the model, it's a permission. Whether an agent runs as the logged-in employee or as its own service-agent user decides everything it can see and do, and a missing field-level permission is why 'the agent can't update the record.' Here's the running-user model, the CRUD/FLS/sharing it inherits, the guest-user surface, and how to scope an agent to least privilege.

  AgentforceSecurityPlatform
• 
  Agentforce Naresh Aug 17, 2026 12 min read

  Agentforce Contact Center: what changes when Salesforce owns the phone line, not just the CRM

  Salesforce spent years being the CRM your contact center integrated with. Agentforce Contact Center is the bet that the integration itself was the problem. Here's what the native CCaaS is, the stack underneath it, the 'routing to resolution' shift, and the honest call on when to go native versus keep your best-of-breed platform.

  AgentforceService CloudContact Center
• 
  AI Agents Naresh Aug 16, 2026 13 min read

  AI agents for wealth management: where the agent preps the meeting, and where the advisor still gives the advice

  Salesforce's Agentic Advisor is aimed squarely at the advisor's busywork, meeting prep, notes, next steps, not the advice itself. That line is the whole design. Here's what the agent can do for a wealth advisor, the FSC data it has to be grounded on, and the compliance boundary: Reg BI, recordkeeping, the AI-notetaker problem, that keeps the recommendation with a licensed human.

  AI AgentsFinancial ServicesAgentforce
• 
  Agentforce Naresh Aug 16, 2026 12 min read

  Testing Agentforce agents in CI/CD: AiEvaluationDefinition, the CLI, and the pipeline the Testing Center can't give you

  The Testing Center is where you learn whether an agent works. It is not where you keep it working. This is the metadata-as-code path: authoring tests as AiEvaluationDefinition, running them headless with sf agent test run, and gating a deploy on the results through the ai-evaluations Connect API, so a regression breaks the build instead of a customer's conversation.

  AgentforceDevOpsTesting
• 
  AI Agents Naresh Aug 16, 2026 12 min read

  AI agents for loyalty program servicing: crediting the missing points without giving away the program

  The highest-volume loyalty ticket is the same everywhere, 'my points didn't post.' It's repetitive, rule-bound, and it hits a system of record that's also a booked liability. That's an ideal agent job and a dangerous one. Here's the servicing workflow an Agentforce agent can own on Salesforce Loyalty Management, the ledger mechanics it has to respect, and the adjustment it should never make on its own.

  AI AgentsLoyaltyAgentforce
• 
  Data 360 Naresh Aug 16, 2026 11 min read

  Reporting on Data 360 with native Salesforce reports: the DMO report types, the join you can't make, and the access nobody enabled

  You unified everything into Data 360 and now someone in ops just wants a report on it, in the report builder they already know, not a SQL client. Native Reports & Dashboards on Data 360 do exactly that, with auto-generated report types for every mapped DMO. Here's how they work, the Calculated Insight join that isn't there yet, the enablement step that trips everyone, and where a report is the wrong tool.

  Data 360ReportingAnalytics
• 
  AI Agents Naresh Aug 15, 2026 11 min read

  Agentforce vs. Amazon Bedrock AgentCore: two opposite bets on the enterprise agent

  They went GA the same day and they are barely the same category. Agentforce sells you an agent already grounded in your CRM; AgentCore sells you the runtime to build one yourself, on any framework. Here's the stack-honest comparison, including where Amazon Q fits, what each one refuses to do, and why a lot of enterprises will end up running both.

  AI AgentsAgentforceAWS
• 
  AI Agents Naresh Aug 15, 2026 13 min read

  AI agents for loan origination: assembling the file without making the credit decision

  A loan application is a document-gathering marathon that ends in a decision the law says a human, or an explainable model, has to own. That split is the whole design. Here's the origination work an agent can take off a loan officer's desk, the adverse-action and fair-lending rules it must not touch, and the line where the credit decision stays a decision, not an inference.

  AI AgentsFinancial ServicesAgentforce
• 
  AI Agents Naresh Aug 15, 2026 11 min read

  MuleSoft Agent Fabric: governing the agents you didn't build

  Your Agentforce agents are the easy part. The hard part is the Bedrock agent finance stood up, the Copilot Studio bot in ops, and the MCP server a developer wired in last quarter: none of which your Salesforce governance can see. Agent Fabric is Salesforce's answer to that sprawl. Here's what it is, the four pieces that matter, and where it earns its place versus where it's still a roadmap.

  AI AgentsMuleSoftGovernance
• 
  AI Agents Naresh Aug 13, 2026 11 min read

  AI agents for RFP and proposal response: where the agent drafts the 200 questions, and where a human still signs the security answers

  An enterprise RFP is the same 200 questions your team has answered a dozen times, re-asked in a new spreadsheet. It's a perfect shape for an agent: high-volume, repetitive, grounded in answers you already own, right up to the questions where a wrong answer is a contractual liability. Here's the response workflow an agent can own, how to ground it on your past proposals without it inventing a compliance claim, and the answer classes that stay human.

  AI AgentsEnterprise AutomationAgentforce
• 
  Data 360 Naresh Aug 13, 2026 11 min read

  Segment Intelligence in Data 360: closing the loop between the audience you activated and the revenue it drove

  Building a segment and pushing it to Meta is the easy half. The half that decides your budget is whether that audience converted, and most teams can't answer it because the engagement data lives in the ad platform, not the profile. Segment Intelligence is the layer that pulls performance back into Data 360 and ties it to the segment that produced it. Here's what it measures, how the round-trip works, and where the credit math bites.

  Data 360SegmentationMarketing
• 
  Agentforce & AI Naresh Aug 13, 2026 11 min read

  Einstein for Service: the assistive AI most teams should deploy before an autonomous agent

  The Agentforce conversation jumps straight to a service agent that resolves the case on its own. But there's a whole assistive layer underneath it: Service Replies, Work Summaries, Case Classification, where the rep stays in control and the AI drafts, summarizes, and classifies. It's faster to ship, lower-risk, and it builds the data foundation the autonomous agent needs. Here's what each tool does, how grounding works, and the licensing maze underneath.

  AgentforceService CloudEinstein
• 
  Agentforce Naresh Aug 13, 2026 12 min read

  Agentforce agent memory: why your agent forgets, and what 'memory' means today

  Everyone wants their agent to 'remember.' But there are three different things hiding under that one word, session state, grounding, and durable cross-session memory, and only two of them are features you can configure right now. Here's what carries context today, where Salesforce is heading, and how to stop building on a capability that hasn't shipped.

  AgentforceMemoryArchitecture
• 
  Agentforce Naresh Aug 13, 2026 13 min read

  Agentforce guardrails: the layered controls that keep an agent in scope

  'The agent hallucinated' is almost never a model problem, it's a guardrails problem. Salesforce draws a hard line between platform guardrails it runs for you and behavioral guardrails you have to build. Here are the five layers that bound what an agent will do, why natural-language instructions aren't enough on their own, and how to test that the boundaries hold.

  AgentforceGuardrailsAI governance
• 
  Data 360 Naresh Aug 13, 2026 13 min read

  Querying Data 360 with SOQL: the SET OPTIONS clause, dataspaces, and the limits that bite

  For years, reading unified data back out of Data Cloud from Apex meant the Query API or ConnectApi and a lot of typecasting. Native SOQL changed that, and Summer '26 added a SET OPTIONS clause that finally lets SOQL reach Data Lake Objects with dataspace scoping. Here's the exact syntax, the NULL-versus-empty-string trap, the callout-and-credit semantics that surprise people, and the limits that send you back to the Query API.

  Data 360SOQLApex
• 
  Agentforce Naresh Aug 12, 2026 15 min read

  How to cut your Agentforce bill: a Flex Credit optimization playbook

  Two orgs running the same agent can pay $0.10 and $0.30 for the identical interaction, the difference is how the agent was built. Here's where Flex Credit burn concentrates, and eight levers to bring it back down without dumbing the agent down.

  AgentforceFlex CreditsCost control
• 
  AI Agents Naresh Aug 12, 2026 13 min read

  AI agents for airline disruption: re-accommodation that doesn't strand a passenger

  When a flight cancels, the cost isn't the cancellation, it's the hours between the schedule breaking and every affected passenger getting rebooked. That gap is where an agent earns its keep. Here's the IRROPS re-accommodation workflow an agent can own, the regulated entitlements it must get right, and the honest line where the airline's PSS and a human take over.

  AI AgentsTravelAgentforce
• 
  AI Agents Naresh Aug 11, 2026 13 min read

  AI agents for KYC and customer onboarding: where the agent assembles the case, and where a human still decides

  An agent can collect the documents, run the screening, untangle the ownership, and hand a compliance officer a finished case in minutes instead of days. What it can't do is sign off the risk rating or file the SAR. Here's the onboarding workflow an agent takes over, the regulatory line it can't cross, and the Financial Services Cloud architecture that keeps the whole thing auditable.

  AI AgentsFinancial ServicesCompliance
• 
  AI Agents Naresh Aug 11, 2026 11 min read

  AI agents for the financial close: faster reconciliation without breaking the audit trail

  The month-end close is a recurring fire drill of reconciliations, variance chasing, and manual journal entries under a hard deadline. It's also one of the best-shaped finance processes for an agent, if you keep the controls intact. Here's the workflow an agent can take over, the SOX boundary it must not cross, and where Salesforce fits when the ledger lives in the ERP.

  AI AgentsFinanceAutomation
• 
  AI Agents Naresh Aug 11, 2026 10 min read

  AI agents for insurance underwriting: submission triage without the black box

  Claims automation gets the attention. The faster underwriting win is the front door: broker submission intake, appetite matching, and clearance that underwriters do by hand on every account. Here's the workflow an agent can take over, where the underwriter's judgment stays, and the grounding that keeps it from binding a risk you don't want.

  AI AgentsInsuranceUnderwriting
• 
  Data 360 Naresh Aug 11, 2026 11 min read

  Headless Data 360: driving your unified data from Claude with the MCP server

  Summer '26 made Salesforce-hosted MCP servers generally available, and there's an open-source Data 360 MCP server in developer preview that fronts ~200 API operations behind three tools. Here's what each one is, the facade trick that keeps a coding agent from drowning, the auth flow, and the read/write line you draw before you point Claude at production data.

  Data 360MCPDeveloper tools
• 
  Platform Naresh Aug 9, 2026 9 min read

  Agentforce for Flow: what the plain-English flow builder drafts, and what it gets wrong

  Describe an automation in a sentence and watch Flow Builder draft it, record-triggered, screen, scheduled. It's the biggest change to admin work in years, and it doesn't burn a single AI credit. But it drafts like a fast, confident junior admin: brilliant on the common path, wrong on nested logic and your pre-2018 fields. Here's how it works and where to keep your hands on the wheel.

  AgentforceFlowAutomation
• 
  AI Agents Naresh Aug 9, 2026 9 min read

  AI agents for channel partners: automating deal registration and MDF in Partner Cloud without leaking one partner's pipeline to another

  Your channel team spends its week answering the same partner questions and chasing deal-registration approvals, while partners abandon the portal because it's faster to just email their rep. An Agentforce agent fixes the drudgery, but a partner-facing agent that gets sharing wrong will show Partner A the deal Partner B just registered. Here's the build, and the boundary that can't move.

  AgentforcePartner CloudPRM
• 
  Agentforce Naresh Aug 9, 2026 11 min read

  Giving Agentforce the open web: the Search the Web action, and the questions it should never answer from memory

  Your agent grounds beautifully on your knowledge base and then falls over the moment a customer asks about something that changed last week. The Search the Web action closes that gap, but a live web result is the least governed thing your agent will ever say. Here's how it works, how it routes through the Trust Layer, and where to fence it.

  AgentforceGroundingWeb Search
• 
  Agentforce Naresh Aug 8, 2026 10 min read

  Agentforce vs. Google Gemini Enterprise: two bets on the agentic enterprise, and when each one wins

  Both promise a fleet of AI agents that act across your business. But one is grounded in your CRM and priced per conversation, the other is grounded in your Google workspace and priced by consumption, and in April 2026 the two vendors agreed to let their agents call each other. Here's the architecture, the pricing math, and the decision rule.

  AgentforceComparisonAI Agents
• 
  AI Agents Naresh Aug 8, 2026 10 min read

  AI agents for healthcare denials and appeals: working the 835 without inventing a diagnosis

  Prior authorization is the fight before the service. Denials are the fight after: a denied claim, a cryptic CARC code, and a 40-day appeals grind that bleeds revenue a health system already earned. An agent can triage and draft the appeal, but only if it reads the remittance correctly and never fabricates clinical fact. Here's the architecture, the codes it has to understand, and the line a human still holds.

  AI AgentsHealthcareRevenue Cycle
• 
  Platform Naresh Aug 8, 2026 11 min read

  Salesforce data archiving: Big Objects, Data Cloud, and the delete decision nobody wants to make

  Your org is 40 GB over its data allocation, the overage invoice is real, and nobody will sign off on deleting anything. Big Objects, Data Cloud, and external archives each solve a different piece of that, and the one thing they all fail at is the reporting you'll want the day after you archive. Here's how each option queries, what it costs, and the retention rule that keeps you out of trouble.

  Big ObjectsData ArchivingStorage
• 
  Agentforce Naresh Aug 8, 2026 11 min read

  Handing an Agentforce conversation to a human: the Omni-Channel flow, the escalation topic, and the context you can't drop

  The demo that resolves a case end to end is easy. The hard, unglamorous part is the moment the agent gives up gracefully: routing the customer to the right human with the transcript, the sentiment, and the case intact. Here's how the handoff works, why the outbound Omni-Channel flow is the real build, and the failure modes that turn escalation into a second wait in line.

  AgentforceService CloudOmni-Channel
• 
  AI Agents Naresh Aug 8, 2026 11 min read

  AI agents for property management: leasing, maintenance triage, and the decision the agent doesn't get to make

  Leasing inquiries and maintenance triage are the two jobs where an agent pays for itself in multifamily. Tenant screening is the one where it can hand you a Fair Housing lawsuit. Here's the architecture that grounds a property-management agent in the PMS, and the bright line: screening, adverse decisions, steering, the agent never crosses alone.

  AI AgentsReal EstateService Cloud
• 
  Data 360 Naresh Aug 8, 2026 11 min read

  Retrievers in Data 360: the grounding layer that decides what your agent reads

  Everyone talks about the search index; the retriever is the part that decides which chunks reach the model. Here's what a retriever is in Data 360, the individual-vs-ensemble choice, the filters and result count that make or break an answer, and why the Data Library built one for you.

  Data 360AgentforceRAG
• 
  Agentforce Naresh Aug 7, 2026 12 min read

  Agentforce Service Agent on email: the highest-volume channel, and the one it's easiest to get wrong

  Web chat, WhatsApp, and voice get the Agentforce demos. Email-to-Case is where most support volume lives, and the email channel behaves differently enough that a messaging-channel mental model will break it. Here's the setup, the ownership rule that decides everything, and the limits nobody mentions until go-live.

  AgentforceService CloudEmail
• 
  Agentforce Naresh Aug 7, 2026 11 min read

  Which model should your Agentforce agent run on? The model menu, the per-subagent override, and how to choose

  Agentforce no longer runs on one model, you can pick a provider, and as of 2026 you can pick a different model for each subagent. That flexibility is useful and a easy way to break a working agent. Here's what's on the menu, what the Salesforce default really is, the standard-actions catch, and a decision framework that isn't 'use the newest one.'

  AgentforceLLMEinstein Trust Layer
• 
  Data 360 Naresh Aug 7, 2026 11 min read

  Connecting Data 360 to Google BigQuery: the zero-copy connector, the auth choice, and the bidirectional bridge

  The BigQuery connector is GA, and it's the cleanest zero-copy path Data 360 has to a warehouse most enterprises already run. But the setup hides two decisions that shape everything after: how Salesforce authenticates to Google, and which direction the data flows. Here's the connector in full, the service-account permissions nobody documents until a query fails, and when to reach for Iceberg instead.

  Data 360Zero CopyIntegration
• 
  Agentforce Naresh Aug 5, 2026 11 min read

  The new Agentforce Builder: what changed now that the legacy builder can't make new agents

  Since the week of July 13 2026, the legacy Agent Builder in Setup can no longer create new agents, every new one is authored in Agentforce Builder in Agentforce Studio, on top of Agent Script. Here's what the new builder is, how the Canvas, Script view, and Preview fit together, what an AiAuthoringBundle is, and how to migrate a legacy agent without breaking the one already in production.

  AgentforceAgent ScriptBuilder
• 
  Data 360 Naresh Aug 5, 2026 10 min read

  Ingesting unstructured data into Data 360: the UDLO and UDMO pipeline underneath Data Library

  A Data Library or Intelligent Context hides it, but every agent grounded on your PDFs and contracts is really running on two objects most teams never look at: the UDLO that references your files and the UDMO that models them. Here's the actual ingestion pipeline: what gets copied and what only gets referenced, how chunking and the search index get wired, and how connecting a UDMO to a structured DMO builds the semantic graph that grounds a multimodal agent.

  Data 360Unstructured dataRAG
• 
  Platform Naresh Aug 5, 2026 11 min read

  Salesforce Data Mask & Seed: safe, realistic sandbox data for testing agents and Apex

  Your sandbox has a full copy of production, which means it has a full copy of your customers' PII, sitting in a lower-security environment where developers, contractors, and now AI agents all poke at it. Here's how Data Mask obfuscates that data in place, what the four masking types do, how seeding gives you realistic data without the risk, and why testing an agent on masked data is a different job than testing Apex.

  Data MaskSandboxesSecurity
• 
  Agentforce Naresh Aug 5, 2026 10 min read

  An Agentforce agent for your pipeline: what it can do about the forecast, and what stays the human's call

  A sales-ops agent that chases stale opportunities and drafts the deal-risk summary is a genuine time-saver. An agent that "manages the forecast" is a category error. Here's the honest split between the agent's job, Einstein's predictive job, and the RevOps human's judgment, plus how to build the useful half without the agent ever touching the number.

  AgentforceSales CloudRevOps
• 
  Agentforce Naresh Aug 5, 2026 10 min read

  Agentforce or build your own agents? The build-vs-buy math that decides it

  The honest question isn't "which is better", it's "where does the cost line cross." Buying Agentforce means a managed platform and consumption pricing; building your own means owning the LLM, the orchestration, the vector store, and the on-call pager. Here's the decision framework, the real TCO components both sides hide, and the workloads where each one wins.

  AgentforceAI StrategyBuild vs Buy
• 
  Data 360 Naresh Aug 4, 2026 11 min read

  Data lineage in Data 360: tracing an agent's answer back to its source

  When an agent quotes a number, someone eventually asks where it came from, and "the model said so" is not an answer. Here's how lineage works in Data 360, what the Informatica acquisition adds on top of the native tracking, and how to use it to prove an agent's grounding instead of hoping it's right.

  Data 360GovernanceData lineage
• 
  Agentforce Naresh Aug 3, 2026 9 min read

  The Agentforce Specialist certification: what the exam tests, and the gap between passing it and shipping an agent

  It's the fastest-growing Salesforce credential, and the study-guide sites will get you a passing score without teaching you to build an agent that survives production. Here's the honest exam blueprint, the platform knowledge people underestimate and fail on, and where the certification stops and the real skill starts.

  AgentforceCertificationCareers
• 
  AI Agents Naresh Aug 3, 2026 10 min read

  AI agents for chargebacks and disputes: building the representment agent when the evidence lives outside Salesforce

  Dispute volume is climbing, the representment window is days, and each case is a deadline-bound evidence hunt across half a dozen systems, the exact shape an agent is good at. Here's the representment architecture on a Salesforce stack, the reason-code and Visa CE 3.0 logic that decides win rates, and the honest truth that the evidence and the rules live in the processor, not the CRM.

  AI AgentsFinancial ServicesPayments
• 
  Data 360 Naresh Aug 3, 2026 10 min read

  Showing Data 360 data on a Salesforce record page: related lists, copy fields, and the mapping that makes it work

  You unified the profile in Data 360, and the service rep looking at the Contact still sees none of it, because reps live on the record page, not in Data Cloud. Here are the two native ways to surface unified data on a CRM record, the identity mapping both depend on, and when a related list is the wrong tool and a data graph is the right one.

  Data 360CRMArchitecture
• 
  Agentforce Naresh Aug 3, 2026 11 min read

  Salesforce Agent Skills: the open format that makes a coding agent build the way your org does

  Salesforce open-sourced a library of Agent Skills, an open, portable format for teaching a coding agent your Apex, LWC, Flow, and Data 360 conventions. It's the same SKILL.md format Claude Code and Cursor use, not a Salesforce-only invention. Here's what a skill is, how progressive disclosure keeps the token cost near zero, and the line between a Skill, an MCP server, and an Agentforce Action that everyone blurs.

  Agentforce VibesAgent SkillsDeveloper Experience
• 
  AI Agents Naresh Aug 3, 2026 12 min read

  AI agents for hotels: building the guest-service and concierge agent when the PMS is the whole build

  A hotel guest agent that answers FAQs is a demo. One that reads a live folio, redeems loyalty points, and modifies a booking is a product, and the hard part isn't the agent, it's the integration into a PMS like OPERA and a loyalty system that were never designed to share a profile. Here's the Salesforce architecture, the read-versus-write boundary that decides your risk, and where a human still has to approve.

  AI AgentsHospitalityAgentforce
• 
  Data 360 Naresh Aug 3, 2026 15 min read

  Data 360 Code Extension: writing Python for transforms and search-index chunking when clicks run out

  The visual transform builder and the declarative chunking config cover most of a Data 360 build, until they don't. Code Extension lets you drop into Python for a Batch Data Transform step or a custom search-index chunking strategy, authored and tested in a notebook. Here's what it can do, where the Python runs, how it differs from the Python Connector, and the limits worth knowing before you reach for code.

  Data 360PythonData Transforms
• 
  Agentforce Naresh Aug 2, 2026 12 min read

  Agentforce triggered agents: building an agent that acts on an event, not a prompt

  Summer '26 made Agentforce agents event-driven, they can now fire on a record change or a Data 360 signal with no human in the loop. That's powerful and expensive in equal measure. Here's the mechanism, the reasoning-vs-flow decision, and the credit math that punishes an over-eager trigger.

  AgentforceAutomationEvent-driven
• 
  Data 360 Naresh Aug 2, 2026 14 min read

  Data 360 governance: making sure your agent only sees what the user is allowed to see

  An Agentforce agent is exactly as trustworthy as the identity it runs as. Here's how object, record, and field access, plus attribute-based policies and dynamic masking: decide what a grounded agent can retrieve, and the retrieval gap that leaks a record no human would ever see.

  Data 360GovernanceAccess control
• 
  AI Agents Naresh Aug 2, 2026 12 min read

  The EU AI Act and your Agentforce deployment: what changed, what applies today, and what Salesforce hands you for free

  The high-risk deadline everyone circled, 2 August 2026, just moved to December 2027 under the Digital Omnibus. But the transparency obligations did not move, and they bind your agent today. Here's the current timeline, mapped article by article onto what you configure in Salesforce and what stays your job.

  ComplianceGovernanceAgentforce
• 
  Agentforce Naresh Aug 1, 2026 11 min read

  Agentforce vs ServiceNow AI Agents: which platform to standardize your digital labor on

  Salesforce and ServiceNow are now selling the same thing, teams of autonomous agents that reason, act, and hand off. The decision isn't which is 'better'; it's which system of record your agents should live closest to, and how each one meters the work. Here's the honest comparison.

  AgentforceServiceNowComparison
• 
  AI Agents Naresh Aug 1, 2026 12 min read

  AI agents for prior authorization: automating the paperwork without automating the denial

  Prior auth is the perfect agent problem, high-volume, rule-bound, fax-and-portal busywork, sitting next to a regulated cliff. An agent can assemble and submit the request; it must not decide medical necessity. Here's the line CMS draws, and the Health Cloud architecture that keeps the agent on the safe side of it.

  AI AgentsHealthcareAgentforce
• 
  Agentforce Naresh Aug 1, 2026 12 min read

  Debugging Agentforce in production: the failures that break agents, and how to see them

  An Agentforce agent that worked in the demo fails without warning in production, a misrouted topic, an action that returns nothing, a permission the Agent User doesn't have. Here's the field guide: the reasoning tools that show you what the agent did, the error catalog by layer, and the permission trap that passes every test and fails on go-live.

  AgentforceTroubleshootingObservability
• 
  Data 360 Naresh Jul 30, 2026 10 min read

  Right to be forgotten in Data 360: the Consent API, and why deleting a profile is harder than it looks

  A GDPR erasure request arrives, and the record you have to delete has been unified across ten source systems into one Data 360 profile. Here's how the Consent API deletion works, the related-DMO rule that decides what survives, the 30/60/90-day reprocessing nobody plans for, and why a 'forgotten' profile can come back.

  Data 360PrivacyCompliance
• 
  Platform Naresh Jul 30, 2026 11 min read

  The Salesforce Pub/Sub API: gRPC event streaming that replaces CometD

  Your CometD subscriber still works, but it is a legacy long-polling client on an API Salesforce no longer invests in. The Pub/Sub API is the gRPC replacement: one interface for publish, subscribe, and schema, with flow control and 72-hour replay. Here is how it works, the Avro decode step everyone forgets, and the limits that bite in production.

  PlatformIntegrationEvents
• 
  Agentforce Naresh Jul 30, 2026 10 min read

  Grounding Agentforce on SharePoint and Confluence: Unified Knowledge is retiring, and what replaces it

  Your agent needs to answer from the knowledge that lives in SharePoint, Confluence, and Google Drive, not in Salesforce. Unified Knowledge was the bridge, and it's scheduled to retire in Summer '27. The successor, Enterprise Knowledge powered by Data 360, works differently. Here's how external-knowledge grounding works, and how to move without stranding your agent.

  AgentforceKnowledgeData 360
• 
  Data 360 Naresh Jul 30, 2026 10 min read

  Data Kits in Data 360: how to move Data Cloud config from sandbox to production

  Everyone hits it eventually: you built the data streams, the DMOs, the calculated insights, and the segments in a sandbox, and there's no Change Set that carries them to production. Data Kits are the answer, and they come in two flavors people constantly conflate. Here's what a data kit packages, the DevOps kit versus the 2GP kit, and the limitations that decide your deployment plan.

  Data 360DevOpsDeployment
• 
  AI Agents Naresh Jul 29, 2026 11 min read

  AI agents for AML alert triage: automating the queue without automating the crime

  Nasdaq Verafin, FIS with Anthropic, and a wave of specialists are shipping agentic AML analysts in 2026, because screening false-positive rates north of 90% have made the alert queue the industry's biggest cost center. Here's the architecture that safely automates L1 triage, why the SAR filing stays human, and where Data 360 and Agentforce fit versus the monitoring engine they don't replace.

  AI AgentsFinancial ServicesCompliance
• 
  Platform Naresh Jul 29, 2026 10 min read

  Flow HTTP Callout: calling a REST API from Salesforce without Apex or middleware

  There's a whole class of integration, check a credit score, validate an address, pull a shipping quote, that used to mean an Apex callout class or a MuleSoft flow. HTTP Callout in Flow does it declaratively: you paste a sample response, Salesforce generates the types, and an admin ships the integration. Here's how it works, the transaction rule that trips everyone up, and the exact line where you still reach for Apex.

  PlatformIntegrationFlow
• 
  Data 360 Naresh Jul 28, 2026 11 min read

  Data 360 vs Adobe Real-Time CDP: the CDP comparison the warehouse debates skip

  The Snowflake and Databricks comparisons answer a different question. This is the one marketers ask: Salesforce Data 360 (formerly Data Cloud) or Adobe Real-Time CDP? Two customer data platforms with opposite centers of gravity: one CRM-native, one experience-native, and the architecture, identity, and activation trade-offs that decide it.

  Data 360AdobeCDP
• 
  Agentforce Naresh Jul 28, 2026 12 min read

  Agentforce for nonprofits: where autonomous agents save staff time

  Salesforce renamed Nonprofit Cloud to Agentforce Nonprofit in December 2025 and shipped four purpose-built agents, Prospect Research, Participant Management, Volunteer Capacity & Coverage, and Donor Support. Here's what each one takes off a small team's plate, the places an agent must never act alone (beneficiaries in crisis is the big one), and the budget math when your org runs on ten donated licenses.

  AgentforceNonprofitUse Cases
• 
  Agentforce Naresh Jul 28, 2026 11 min read

  The Agentforce setup checklist: permissions, topics, and actions to verify before launch

  The build tutorials start at Agent Builder; the failures usually start earlier, in the org. This is the readiness pass to run before anyone builds: licensing and credits, the enablement order, the two permission surfaces, Trust Layer posture, grounding and channel prerequisites, and the testing and governance gates that decide whether launch day is boring.

  AgentforceSetupGo-Live
• 
  Consulting Naresh Jul 28, 2026 11 min read

  The Salesforce implementation checklist: what to confirm before you sign the SOW

  Most Salesforce implementations don't fail in the build phase, they fail at signing, when scope, data ownership, environments, and acceptance criteria were left vague in the SOW. Here's the pre-contract checklist a mid-market buyer should walk through before engaging an implementation partner: eleven things to pin down in writing, and the red flags that predict failure.

  ConsultingImplementationSalesforce
• 
  Agentforce Naresh Jul 27, 2026 11 min read

  Building an Agentforce action from a Flow: the no-code path, and the interface you can't take back

  You don't need Apex to give an agent a new capability. An autolaunched Flow, wired up as an action, is the path most admins should reach for first, but the variable descriptions decide whether the agent ever calls it correctly, and the input/output contract locks the moment you activate. Here's the whole build, the running-user trap, and where a Flow action still loses to Apex.

  AgentforceFlowAgent actions
• 
  AI Agents Naresh Jul 27, 2026 10 min read

  AI agents for supply chain disruption: what Salesforce can do, and what the planning system still owns

  A port strike, a supplier miss, a weather event, the cost isn't the disruption, it's the hours between the signal and someone telling the affected customers. That gap is where an agent earns its keep. But Salesforce is not your planning optimizer, and pretending otherwise is how these projects fail. Here's the honest split: the commercial-response work an agent should own, the planning it should never fake, and the architecture that connects the two.

  AI AgentsSupply chainManufacturing
• 
  Data 360 Naresh Jul 27, 2026 11 min read

  Apache Iceberg in Data 360: file federation, the REST catalog, and the open-table bridge zero copy was missing

  The native zero-copy connectors cover Snowflake and BigQuery. Everything else: your S3 data lake, a governed Iceberg catalog, a table another team already curated, comes in through file federation, and it works at the open-table-format layer instead of a per-vendor connector. Here's what an Iceberg File Federation connection is, how the REST catalog and Lake Formation govern it, and where the query-in-place promise materializes.

  Data 360Apache IcebergZero copy
• 
  Agentforce Naresh Jul 26, 2026 12 min read

  Why your Agentforce agent keeps picking the wrong topic: designing subagents, actions, and classification descriptions

  The refund question routes to the returns subagent every third try, and no amount of instruction-tweaking fixes it. The problem isn't your prompt: it's that two subagents describe overlapping jobs, and the Agent Router can only read what you wrote in the classification. Here's how routing works, how to design subagents that don't collide, and the 15-topic ceiling that forces the next decision.

  AgentforceAgent designAtlas
• 
  AI Agents Naresh Jul 26, 2026 10 min read

  AI agents for recruiting: automating sourcing, screening, and scheduling without failing a bias audit

  A recruiting agent can reclaim the 30 hours a week a recruiter loses to sourcing and scheduling. It can also, if you let it rank and reject candidates, turn your hiring into a regulated high-risk AI system overnight: auditable under NYC Local Law 144 and the EU AI Act. Here's where an agent safely acts, where a human must still decide, and the architecture that keeps the line defensible.

  AI AgentsRecruitingCompliance
• 
  Data 360 Naresh Jul 26, 2026 10 min read

  Keyword, vector, or hybrid: choosing the Data 360 search index that grounds your agent

  Your agent retrieves a confident, fluent, wrong answer, and the culprit is upstream of the model, in how you indexed the data. Vector search misses the exact part number; keyword search misses the paraphrase. Here's what each Data 360 search index type does, how hybrid merges and re-ranks the two, and the retrieval-quality decision that sets both your answer accuracy and your credit bill.

  Data 360RAGSearch index
• 
  Agentforce Naresh Jul 25, 2026 11 min read

  A2A in Agentforce: letting your agent talk to agents you don't own

  MCP taught your agent to use tools. A2A teaches it to delegate to other agents: a logistics agent on Google Cloud, a procurement agent on Azure, that you can't see inside. Here's what the Agent2Agent protocol standardizes, the Agent Card and task lifecycle, how it differs from MCP and Atlas orchestration, and the trust boundary you're opening.

  AgentforceA2AInteroperability
• 
  AI Agents Naresh Jul 25, 2026 11 min read

  AI agents for contract lifecycle management: where the agent drafts, and where a lawyer still signs

  An agent that reads a 60-page MSA and flags the indemnity gap in two minutes is a real win. An agent that confidently cites a clause that isn't in the contract is a malpractice claim. Here's the CLM architecture that gets the speed without the liability: grounding, deterministic extraction, and the human gate you design in from the start.

  AI AgentsLegalContracts
• 
  Data 360 Naresh Jul 25, 2026 11 min read

  Salesforce Data 360 vs. Microsoft Fabric: OneLake, the zero-copy bridge, and the job each one is for

  Fabric and Data 360 both store data, both run SQL, and both now claim agents, so procurement asks which one wins. It's the wrong question. Here's how OneLake shortcuts and Data 360 federation bridge the two without copying a row, where each platform earns its place, and the identity-resolution gap that decides most of it.

  Data 360Microsoft FabricArchitecture
• 
  Agentforce Naresh Jul 24, 2026 11 min read

  Agentforce Revenue Management: where an agent can build the quote, and where a human still signs

  Revenue Cloud is now Agentforce Revenue Management, and the demo shows an agent quoting a deal in a sentence. The moment an agent quotes a price is the moment it can invent one, and in quote-to-cash, a made-up number is a contract you have to honor. Here's the real object model, how to ground the agent so it prices from the engine and never from the model, and the three places a human still has to sign.

  AgentforceRevenue CloudQuote-to-Cash
• 
  AI Agents Naresh Jul 24, 2026 14 min read

  AI agents for customer success: acting on a churn score without letting the agent give away the discount

  A churn model tells you an account is slipping. The hard part is what happens next, and the tempting mistake is to let the agent fix it by offering a credit. Here's the retention architecture that wakes an agent on a real signal, lets it draft the save play, and stops short of the one decision it must never make on its own.

  AI AgentsCustomer SuccessData 360
• 
  Data 360 Naresh Jul 24, 2026 12 min read

  Salesforce Data 360 vs. Databricks: the zero-copy bridge, the shared models, and the day the partner became a competitor

  Data 360 and Databricks look like rivals on a procurement slide and behave like complements in a real architecture, right up until June 2026, when Databricks shipped a CDP of its own. Here's how the bidirectional zero-copy bridge works, where each platform wins, and how to design for both without betting the estate on a partnership that's now also a competition.

  Data 360DatabricksArchitecture
• 
  Agentforce Naresh Jul 23, 2026 10 min read

  Putting an Agentforce agent on your website: Messaging for Web, the snippet, and the pre-chat trap

  You built a Service Agent and it tests beautifully in the preview window, but the chat bubble on your own website is a different deployment entirely. Here's the Messaging for Web build: the channel, the Embedded Service snippet, the domains that 401 you, and the hidden pre-chat field that ships an impersonation hole.

  AgentforceMessagingDeployment
• 
  AI Agents Naresh Jul 23, 2026 10 min read

  AI agents for order management: automating fulfillment exceptions and returns without misplacing an order

  The order that ships and arrives never needs an agent. The value is in the exceptions: the split shipment, the out-of-stock line, the return that has to go back to the right warehouse. Here's the Salesforce Order Management object model an agent acts on, what Agentic Order Management automates, and the two places a human still has to sign.

  AI AgentsOrder ManagementCommerce
• 
  Data 360 Naresh Jul 23, 2026 9 min read

  Advertising Studio is retiring: moving your first-party ad audiences to Data 360

  Marketing Cloud Advertising Studio reaches end of life in 2026, and the replacement isn't a rename, it's Data 360 Ad Audiences, a different data model, a different identity layer, and a different consent story. Here's what changes, the match-rate reality nobody warns you about, and a migration sequence that fits the deadline.

  Data 360AdvertisingActivation
• 
  Agentforce Naresh Jul 22, 2026 11 min read

  Agentforce Grid: running AI on ten thousand records without writing a batch job

  A single prompt that summarizes one case is easy. Running it across 40,000 records used to mean an Apex batch and a prompt-template harness. Agentforce Grid is the spreadsheet-shaped surface Salesforce built for that job. Here is what it is, the three column types, the compounding-workflow trick, and where the credits go.

  AgentforceData 360Automation
• 
  AI Agents Naresh Jul 22, 2026 10 min read

  AI agents for insurance claims: automating FNOL and triage without ceding the adjuster decision

  The claims agent that files a first notice of loss and triages the routine 70% is a genuine near-term win. The one that decides coverage and cuts the check is a compliance incident waiting to happen. Here is the Salesforce Insurance Cloud data model it runs on, the FNOL-to-triage architecture, and the regulatory line you design the human back into.

  AI AgentsInsuranceAgentforce
• 
  Data 360 Naresh Jul 22, 2026 8 min read

  The Data 360 Query API: running SQL against your unified data from outside Salesforce

  Your unified profiles live in Data 360, but the app that needs them lives somewhere else. The Query API is the door: ANSI SQL over the Hyper engine, reachable by any external system. Here is the two-step auth that trips everyone up, the batch pagination model, the 49,999-row wall, and how to not melt your credit budget.

  Data 360APIIntegration
• 
  Data 360 Naresh Jul 22, 2026 13 min read

  The Data 360 Ingestion API: streaming vs bulk, and how to push data in without surprises

  The native connectors cover Salesforce and the big warehouses. Everything else: your billing system, an IoT feed, a custom app, comes in through the Ingestion API. Here's the two patterns, the setup nobody documents well, the token dance that 401s your first call, and the latency and credit math that catch teams out.

  Data 360Ingestion APIIntegration
• 
  Agentforce Naresh Jul 21, 2026 12 min read

  Context and custom variables in Agentforce: how to stop your agent re-asking what it already knows

  Your agent verified the customer, pulled their account, and then, two turns later, asked for their email again. That's not a model problem, it's a variables problem. Here's the difference between context, session, and custom variables, how to pass state deterministically between actions, and the pre-chat trap that ships an impersonation hole.

  AgentforceVariablesArchitecture
• 
  AI Agents Naresh Jul 21, 2026 13 min read

  AI agents for procurement: automating intake, approvals, and sourcing without losing spend control

  The whole S2P suite market pivoted to agents in 2026, SAP, Coupa, Zip, Ivalua, ORO. The prize is intake and approvals, where maverick spend starts and cycle times stall. Here's the reference architecture, the segregation-of-duties problem agents create by construction, and the honest map of where Salesforce fits and where it doesn't.

  AI AgentsProcurementAutomation
• 
  AI Agents Naresh Jul 21, 2026 11 min read

  AI agents for accounts payable: automating invoice processing without losing the audit trail

  The consumer booking agent gets the headlines. The faster, safer enterprise win is accounts payable: invoice capture, three-way match, and exception resolution that finance teams do by hand thousands of times a month. Here is the architecture that works, where the agent decides versus where a human still signs, and the two controls that keep it auditable.

  AI AgentsFinanceAutomation
• 
  Data 360 Naresh Jul 21, 2026 11 min read

  Data Lake Objects vs Data Model Objects in Data 360: the mapping step that decides everything downstream

  Every data stream you create in Data 360 produces a Data Lake Object automatically. Nothing useful happens until you map it into a Data Model Object, and that mapping is where identity resolution, segmentation, and agent grounding either work or break. Here is what each object is, why the mapping exists, and the decisions that bite months later.

  Data 360Data modelingData Cloud
• 
  Agentforce Naresh Jul 21, 2026 11 min read

  Deploying an Agentforce agent from sandbox to production: the metadata map nobody hands you

  You built the agent in a sandbox, it passed testing, and now you have to move it to production, and the change set is missing half of it. An Agentforce agent is not one object; it is a stack of Bot, planner, topic, and action metadata with a strict dependency order. Here is the whole deployment, the ordering that stops it failing, and the pieces that never travel in the metadata at all.

  AgentforceDevOpsMetadata
• 
  Agentforce Naresh Jul 20, 2026 12 min read

  Agentforce Assistant: what Einstein Copilot became, and when to use it instead of an autonomous agent

  Einstein Copilot did not get discontinued, it got renamed twice and reclassified into the reactive, in-app assistant that runs as you. Meanwhile every Agentforce headline is about autonomous agents that run as themselves. Here is the difference that decides which one you build.

  AgentforceEinstein CopilotEmployee agents
• 
  AI Agents Naresh Jul 20, 2026 12 min read

  AI agents for corporate travel and expense: automating T&E without losing policy control

  The consumer booking agent gets the keynote. The faster enterprise ROI is the unglamorous back office: booking within policy, expense reports that fill themselves out, approvals that auto-clear the low-risk 90%. Here is the architecture, what the vendors ship, and the two problems that decide whether it is safe.

  AI AgentsTravelAutomation
• 
  Data 360 Naresh Jul 20, 2026 14 min read

  Data 360 Clean Rooms: collaborating on customer data without moving a single row

  Two brands want to measure a shared audience, and neither legal team will let the other see a raw customer record. A clean room is the join that happens where nobody owns the result. Here is how Data 360 Clean Rooms work, the privacy controls that enforce the promise, and where the credits and limits bite.

  Data 360Clean RoomsData collaboration
• 
  AI Agents Naresh Jul 19, 2026 9 min read

  Collections with Agentforce: predictive invoice scoring and risk-based dunning, explained

  Summer '26 shipped Collections with Agentforce, predictive AI that scores every open invoice, recommends a risk-based dunning plan, and adapts outreach by customer tier. Here's how it sits on Revenue Cloud Billing's real object model, what the agent decides versus what a human still approves, and where the data has to be right first.

  AgentforceRevenue CloudEnterprise Automation
• 
  Platform Naresh Jul 19, 2026 10 min read

  Salesforce Foundations: what's free, and the point where you outgrow it

  Foundations is the $0 SKU that hands Enterprise Edition orgs Agentforce, Data 360, and a slice of every cloud, including 200,000 free Flex Credits. Here's exactly what you get, the credits math, the one setting you can't undo, and how to tell when you've outgrown the free tier.

  Salesforce FoundationsAgentforcePricing
• 
  Platform Naresh Jul 19, 2026 10 min read

  Setup with Agentforce: the admin agent that configures your org, and where it stops

  Setup with Agentforce went GA in Summer '26, an agent that lives inside Salesforce Setup and builds objects, fields, flows, permission sets, and report types from natural language. Here's what it can do, what it can't yet, the credit-cost nuance, and how to turn it on without opening a hole.

  AgentforceSalesforce AdminSetup
• 
  Agentforce Naresh Jul 18, 2026 11 min read

  Agentforce for Marketing: what Marketing Cloud Next automates, and what it doesn't

  Salesforce rebuilt Marketing Cloud on the core platform, put a Campaign Creation Agent in front of it, and rebranded the whole thing Agentforce Marketing. The campaign-from-a-sentence demo is real. The migration, the AMPscript gap, and the credit math are the parts nobody puts on the keynote slide.

  AgentforceMarketing CloudData 360
• 
  Agentforce Naresh Jul 18, 2026 10 min read

  Agentforce IT Service: Salesforce's run at ServiceNow, and where the CMDB story is still thin

  Salesforce shipped an agent-first ITSM product built on Service Cloud, priced it aggressively per seat, and started naming ServiceNow customers it won. The employee-agent story is strong. The CMDB and discovery story: the thing ITSM runs on, is where the honest evaluation lives.

  AgentforceIT ServiceITSM
• 
  Data 360 Naresh Jul 18, 2026 12 min read

  The Data 360 Web SDK: capturing first-party behavior an agent can ground on

  An agent grounded on last night's batch load can't personalize for the customer clicking right now. The Data 360 Web SDK is how you get real-time website behavior into the profile, but the schema, the identity stitching, and the streaming-credit bill are where teams get stuck. Here's the build.

  Data 360Web SDKReal-time data
• 
  Platform Naresh Jul 17, 2026 11 min read

  The Salesforce DX MCP Server: driving your org from Claude, Cursor, and VS Code

  Salesforce shipped an official MCP server that lets any MCP-capable AI client deploy metadata, run Apex tests, query with SOQL, and spin up scratch orgs, without you pasting credentials into a prompt. Here's what @salesforce/mcp exposes, how the org allowlist keeps it from touching production by accident, and where the NON-GA tools bite.

  MCPSalesforce DXDeveloper tools
• 
  Data 360 Naresh Jul 17, 2026 10 min read

  Tableau Next: what changes when your BI platform is an agent, not a dashboard

  Salesforce is pitching Tableau Next as the first agentic analytics platform, built on Data 360, driven by a semantic model, and staffed by skills that prep data, answer questions in natural language, and watch metrics for you. Here's what's real, how it sits on the semantic layer you may already have, and the licensing shift that catches finance off guard.

  Tableau NextAnalyticsData 360
• 
  Agentforce Naresh Jul 15, 2026 13 min read

  Deploying an Agentforce agent to WhatsApp and SMS: the routing flow is the whole build

  Your customers live in WhatsApp and SMS, not the chat bubble on your website. Putting an Agentforce Service Agent there isn't a channel toggle: it's an Omni-Channel Flow that routes the conversation, a WhatsApp Business Account with a 24-hour window, and a handoff most teams forget to build. Here's the whole deployment, and the three meters it bills against.

  AgentforceWhatsAppMessaging
• 
  Data 360 Naresh Jul 15, 2026 13 min read

  Predictive AI in Data 360: native models, bring-your-own-model, and getting a score to do something

  Everyone's building generative agents; the quiet workhorse is still a churn score or a propensity model. Data 360 can train one natively or call your SageMaker, Vertex, or Databricks model in place. Here's the honest map: what native AutoML can and can't do, what 'zero-copy' really means at inference, why the Trust Layer doesn't cover any of it, and how a prediction becomes an action.

  Data 360Predictive AIBYOM
• 
  Platform Naresh Jul 15, 2026 12 min read

  Salesforce Flow Orchestration in 2026: the multi-step engine that just stopped costing extra

  Flow Orchestration became a standard feature in Spring '26, no add-on, no per-run bundles. That removes the one reason most teams never touched it, right as it becomes the layer that coordinates humans, agents, and systems across a process. Here's the object model, the new Flow-based approvals, where Agentforce plugs in, and when it's overkill.

  FlowAutomationAgentforce
• 
  Agentforce Naresh Jul 15, 2026 12 min read

  Connecting Agentforce to an MCP server: giving an agent external tools without hand-building every action

  You can wire Agentforce to Snowflake, Jira, GitHub, or your own service by writing a custom Apex action for each one, or you can register an MCP server once and let the agent discover its tools. Here's how Agentforce's MCP client support works, how the Agentforce Registry and API Catalog split governance, and the beta caveats and trust boundaries that decide whether it's safe to point at production.

  AgentforceMCPIntegration
• 
  Data 360 Naresh Jul 15, 2026 11 min read

  Data Cloud One: how to give five Salesforce orgs one Data 360 without five bills

  Enterprises rarely run one org. They run a sales org, a service org, a couple of acquisitions, and a regional instance, and each one wants unified customer data. Standing up a separate Data 360 in every org is the expensive wrong answer. Here's how Data Cloud One connects companion orgs to a single home org, what those companion orgs can and can't do, and where the licensing and SDLC edges are.

  Data 360Multi-OrgArchitecture
• 
  Data 360 Naresh Jul 13, 2026 12 min read

  Data Transforms in Data 360: batch vs. streaming, and shaping data before it hits your model

  Ingested data almost never arrives in the shape your model needs. Data Transforms are where you join, clean, and reshape it before it becomes a unified profile, and the batch-vs-streaming choice decides whether you can join at all. Here's what each does, the SQL and the nodes, and the streaming gotcha that locks your output out of segmentation.

  Data 360Data TransformsSQL
• 
  Data 360 Naresh Jul 13, 2026 11 min read

  Segmentation and activation in Data 360: building an audience that reaches the target

  Building a segment is the easy part. Getting the right people, with the right contact points, into the right platform, and keeping the count honest: is where Data 360 audiences go wrong. Here's how segmentation and activation work, the related-attribute rule that decides what you can send, and why your segment count never matches Data Explorer.

  Data 360SegmentationActivation
• 
  Agentforce Naresh Jul 13, 2026 12 min read

  MuleSoft for Agentforce: turning your APIs into governed agent actions

  Your agent needs to touch the systems that run your business, the order platform, the billing engine, the legacy service your APIs already front. You can wire each one up by hand in Apex, or you can publish it once as a MuleSoft topic and let every agent reuse it. Here's what Topic Center and the API Catalog do, how a Mule API becomes an Agentforce action, and where the governance lives.

  MuleSoftAgentforceIntegration
• 
  Agentforce Naresh Jul 13, 2026 11 min read

  Agentforce Data Library: grounding an agent on your knowledge and PDFs without hand-building RAG

  You want your agent to answer from your Knowledge base and your PDFs, and every guide points you at a full Data Cloud RAG build. A Data Library is the shortcut: it auto-provisions the same search index and retriever with the tuning hidden and set to defaults. Here's what it creates, when the defaults are enough, and the moment you have to graduate to a hand-built pipeline.

  AgentforceRAGData Library
• 
  Agentforce Naresh Jul 13, 2026 10 min read

  Agentforce for Field Service: where an agent can touch the schedule, and where a dispatcher still has to

  Salesforce put agents inside Field Service, some that book and rebook appointments with no human involved, some that only hand a dispatcher options to choose from. The line between those two modes is the whole design. Here's how the agent plugs into the scheduler, what runs autonomously, and where a human still has to stand.

  AgentforceField ServiceScheduling
• 
  Data 360 Naresh Jul 13, 2026 12 min read

  Data 360 automation: Data Cloud-triggered flows vs. Data Actions vs. activations

  You have a signal in Data 360 and you want something to happen, a task created, a webhook fired, a journey started. There are four surfaces that can do it, they behave nothing alike, and picking the wrong one buys you either seconds of latency you didn't need or minutes you couldn't afford. Here's what each mechanism does and the decision rule.

  Data 360FlowAutomation
• 
  Agentforce Naresh Jul 13, 2026 10 min read

  Synthetic personas: how to test an Agentforce agent against customers who don't exist

  Agentforce Testing Center can now simulate whole conversations with synthetic personas, a frustrated customer, a non-native English speaker, and gate your pipeline on the result. Useful, until you learn what the UX-research field already knows: synthetic personas cluster around the average customer and miss the edge cases that break agents. Here's how to use them without shipping an agent that only works for people who don't exist.

  AgentforceTestingSynthetic personas
• 
  Agentforce Naresh Jul 12, 2026 12 min read

  Agentforce Voice: what changes when the agent has to answer the phone

  Voice isn't a new agent, it's a channel in front of the one you already built. But a channel with no buttons, a sub-second latency budget, and a caller who talks over you rewrites how you design topics, confirmations, and the handoff to a human. Here's how Agentforce Voice works and where the sharp edges are.

  AgentforceVoiceService Cloud
• 
  Agentforce Naresh Jul 11, 2026 12 min read

  Custom Apex actions for Agentforce: the description is the code that matters most

  You annotate a method, wire it into Agentforce Builder, and the agent still won't call it, or calls it with garbage. Custom Apex actions are how an agent does anything standard actions can't, but the part that decides whether they work isn't the Apex. It's the description Atlas reads, the permission set the agent user is missing, and the governor limits a reasoning loop can blow through. Here's the whole build.

  AgentforceApexAgent Actions
• 
  Data 360 Naresh Jul 11, 2026 11 min read

  Calculated vs. streaming insights in Data 360: the engine choice that sets your credit bill

  Both compute a metric from your data; they are not interchangeable. One batches the whole history on a schedule, the other aggregates a live event stream inside a time window, and picking the wrong one is how a metric nobody reads by morning ends up costing many times what it should. Here's the SQL, the window clause, and the decision rule.

  Data 360Calculated InsightsSQL
• 
  Consulting Naresh Jul 11, 2026 10 min read

  Agentforce implementation partner: what to look for in 2026

  The two firms mid-market buyers called first for Agentforce, NeuraFlash and Coastal Cloud, were both acquired into global SIs across 2025–26. Here's what separates a capable Agentforce implementation partner from a generalist Salesforce SI: the questions to ask before you sign, the red flags, and what a real eight-week engagement looks like.

  AgentforceConsultingImplementation
• 
  Agentforce Naresh Jul 11, 2026 7 min read

  Agentforce ROI: how to calculate it honestly before you build (2026)

  Everyone asks what Agentforce ROI looks like; the honest answer is that it depends on inputs you control. Here's the actual model: the cost side (Flex Credits vs licenses), the benefit side (case deflection math), the break-even volume most use cases need, and the soft savings you should refuse to count.

  AgentforceROIPricing
• 
  Agentforce Naresh Jul 11, 2026 11 min read

  Agentforce sales agents: what changes when the agent prospects on its own

  Most teams build a service agent first, then assume the SDR agent is the same pattern pointed at leads. It isn't. Sales agents act unsupervised and outbound, the standard ones are configured rather than coded, and the failure mode is emailing a prospect something wrong. Here's how Agentforce SDR and Sales Coach work, and where a human still has to stand.

  AgentforceSales CloudAI Agents
• 
  Consulting Naresh Jul 11, 2026 11 min read

  How to Choose the Best Salesforce Consulting Firm for a Small or Mid-Sized Business (2026)

  AppExchange lists over 2,000 Salesforce consulting partners, and most of them are built to serve enterprises, not a 40-person company. Here's how a small or mid-sized business vets a firm in 2026: the credentials that mean something, the pricing model that fits, the AI-readiness test, and the red flags that cost you a re-implementation.

  Salesforce ConsultingSMBBuyer's Guide
• 
  Data 360 Naresh Jul 11, 2026 9 min read

  Data spaces in Data 360: the partition that decides who unifies with whom

  A data space looks like a folder for organizing your data. It isn't. It's a hard boundary for identity resolution, access, and activation, and once profiles unify inside one, they don't cross into another. Here's what a data space separates, how the enhanced-security permission model works now, and the one-vs-many decision you can't cleanly undo.

  Data 360Data CloudData Architecture
• 
  AI Agents Naresh Jul 11, 2026 10 min read

  Human-in-the-loop for AI agents: designing the approval gate before you need it

  Full autonomy is the wrong default for any agent action that's hard to reverse. The fix isn't a human rubber-stamping everything: it's classifying actions by blast radius and reversibility, then gating only the consequential ones. Here's how to design approval gates that scale, grounded in what Agentforce, Flow, and Omni-Channel give you.

  AI AgentsAgentforceGovernance
• 
  AI Agents Naresh Jul 9, 2026 8 min read

  Agentforce in Slack: why the only agent that runs there is an Employee Agent, and what that decides

  Slack has become the front door to Agentforce for internal work, but you can only deploy one agent type there, and it runs as the person who invoked it. That single design choice sets your entire permission model, your grounding strategy, and your credit bill. Here's how the integration works and where the sharp edges are.

  AgentforceSlackEmployee Agents
• 
  Agentforce Naresh Jul 9, 2026 10 min read

  Bring your own LLM to Salesforce: the honest map of Model Builder, the Open Connector, and what Agentforce lets you swap

  You can run Salesforce AI on your own OpenAI, Claude, or Gemini account, but the surface that lets you pick a model for Prompt Builder is not the one that picks the model for Agentforce. Here's what BYOLLM, the LLM Open Connector, and Agentforce model selection each do, and how to wire a custom model in without it leaking data.

  BYOLLMEinstein Trust LayerModel selection
• 
  Agentforce Naresh Jul 9, 2026 9 min read

  Migrating Einstein Bots to Agentforce: why it's a rebuild, and how to run it without a big-bang cutover

  There's no upgrade button from Einstein Bots to Agentforce because the two share zero authoring objects, you rebuild. Here's what the beta migration tool generates, the hybrid architecture that lets both run side by side during the transition, and the Omni-Channel Flow handoff that carries context across the seam.

  AgentforceEinstein BotsMigration
• 
  Agentforce Naresh Jul 9, 2026 10 min read

  The Agentforce Agent API: putting a governed agent inside your own app

  You built an agent in Studio, and it's trapped in the Salesforce chat widget. Your customers live in your mobile app; your ops runs in a queue. Headless 360 exposes Agentforce as an API: here's how the Agent API drives an agent from your own code, how it differs from MIAW, and the identity and cost boundaries that decide whether it's safe.

  AgentforceAgent APIIntegration
• 
  Data 360 Naresh Jul 9, 2026 11 min read

  Data graphs in Data 360: the low-latency layer that grounds an agent on a live record

  An agent mid-conversation can't run a 12-table join and make the customer wait. Data graphs are Data 360's answer: a pre-joined, denormalized view of a customer served in milliseconds. Here's how they work, the real-time vs. batch trade-off, the 200 KB discipline nobody explains, and where they cost you.

  Data 360Data graphsAgentforce
• 
  Agentforce Naresh Jul 9, 2026 12 min read

  Salesforce Prompt Builder: grounded prompt templates that don't drift into fiction

  The difference between an AI feature people trust and one they switch off is almost never the model, it's the prompt and what it's grounded on. Prompt Builder is where that gets decided, and most teams treat a data-binding surface like a text box. Here's the template types, the grounding ladder, the Apex, and where it drifts.

  Prompt BuilderAgentforceGrounding
• 
  Agentforce Naresh Jul 8, 2026 11 min read

  Agentforce Vibes pricing and licensing: what changes when the coding agent builds in parallel

  Agentforce Vibes stopped being free for most orgs on June 1, 2026. Here's exactly what it costs now: Flex Credits versus the permission-set license, and the limited free Developer Edition allowance, plus how the agentic coding workflow works and where a human still has to stand.

  Agentforce VibesDeveloperAgentic AI
• 
  Data 360 Naresh Jul 8, 2026 13 min read

  Intelligent Context: getting agents to read your documents without hallucinating the answer

  The semantic layer fixed the metric problem. Intelligent Context is Data 360's answer to the other half: grounding agents on the PDFs, contracts, and transcripts where naive RAG retrieves the wrong chunk. Here's the pipeline, the chunking decisions that make or break it, and where the credits go.

  Data 360Intelligent ContextRAG
• 
  AI Agents Naresh Jul 8, 2026 11 min read

  Agent sprawl is the new shadow IT: governing a fleet before it governs you

  It took a decade to spin up your first agent and six clicks to spin up your tenth. Most orgs now run dozens with no owner, no scoped identity, and no retirement plan, and half are unmonitored. Here's the governance a fleet needs, grounded in what Agentforce gives you: running-user identity, the audit trail, and lifecycle management.

  AI AgentsGovernanceIdentity
• 
  Agentforce Naresh Jul 7, 2026 11 min read

  Agentforce observability: what Command Center shows you, and what it doesn't

  You can't manage a digital workforce you can't see. Agentforce Observability went GA with Command Center, Agent Analytics, and OpenTelemetry session traces. Here's the telemetry you get, how to export it, and the metrics that tell you an agent is failing quietly.

  AgentforceObservabilityOperations
• 
  AI Agents Naresh Jul 7, 2026 14 min read

  Agentic commerce for Salesforce merchants: what to build when the buyer is an agent

  ChatGPT and Gemini now buy on a customer's behalf, and Agentforce Commerce went GA to meet them. The storefront you optimized for a decade is being replaced by two machine-readable surfaces: a product feed and a checkout API. Here's what they are, how delegated payment works, and where Salesforce fits.

  Agentic commerceAgentforce CommerceACPPayments
• 
  AI Agents Naresh Jul 7, 2026 11 min read

  AI agents in travel: building the booking and disruption agent that doesn't invent a fare

  Sabre, PayPal, and Mindtrip shipped end-to-end agentic booking; airlines are automating disruption rebooking. But an agent that quotes a fare it made up is a liability. Here's the data and API architecture a travel brand needs to ground an agent in real inventory, and where Agentforce fits.

  AI AgentsTravelAgentforce
• 
  Data 360 Naresh Jul 7, 2026 12 min read

  Grounding agents on a semantic layer: how Data 360 stops them inventing your metrics

  Point an agent at your raw tables and ask for 'revenue,' and it will confidently make one up. Tableau Semantics turns Data 360 into a governed metric layer agents query instead of guessing. Here's what a semantic model is, how it grounds an agent, and where it stops.

  Data 360Semantic layerAgentforce
• 
  Agentforce Naresh Jul 6, 2026 13 min read

  Agentforce multi-agent orchestration: designing agent teams that survive the handoff

  Multi-agent orchestration went GA in Summer '26, and the routing surface is now an agent description, not a topic. Here's how the orchestrator delegates, where handoffs leak context and credits, and how to design a team that doesn't misroute.

  AgentforceMulti-agentArchitecture
• 
  Agentforce Naresh Jul 5, 2026 11 min read

  How to test an Agentforce agent: from utterance batches to full-conversation simulation

  Agentforce Testing Center went from a CSV-driven, single-turn tool to a Studio tab that simulates whole conversations with synthetic personas and gates your CI/CD pipeline. Here's the 2026 testing stack, layer by layer, and the pre-launch sequence we run before any agent meets a customer.

  Testing CenterAgentforce StudioQA
• 
  Agentforce Naresh Jul 1, 2026 13 min read

  Agentforce 360 for practitioners: what's real, what's GA, and what to adopt first

  Agentforce 360 launched at Dreamforce as “available globally today”, while half its headline features were in pilot. Nine months on, here's the component-by-component status check, and the order we'd adopt it in for a typical service org.

  Agentforce 360Data 360Voice
• 
  Data 360 Naresh Jun 24, 2026 13 min read

  Does Agentforce need Data Cloud? The honest Data 360 dependency map

  Every Agentforce pricing call ends with the same question: do we have to buy Data 360 too? Here's the honest dependency map: what's provisioned free, what silently bills against it, and what requires a paid implementation.

  AgentforceData 360Licensing
• 
  Agentforce Naresh Jun 18, 2026 3 min read

  Five things that decide whether your Agentforce agent works

  Agentforce demos beautifully and fails without warning. After deploying agents in production, here is what separates the ones that deflect real tickets from the ones that embarrass you.

  AgentforceData CloudAI governance
• 
  Org Health Naresh Jun 18, 2026 4 min read

  Why your Salesforce org slows down, and how to pay back technical debt

  Every org accumulates technical debt. Here is how to spot it, prioritise it, and pay it back without freezing delivery for six months.

  Technical debtAutomationRelease management
• 
  AppExchange Naresh Jun 17, 2026 4 min read

  What the AppExchange Security Review checks, and how to pass the first time

  The Security Review is the gate every AppExchange app must clear. Here is what reviewers look for and how to build so you pass on the first submission.

  Security ReviewISVApex
• 
  Automation Naresh Jun 16, 2026 4 min read

  Flow, Apex, or Agentforce? A decision guide for Salesforce automation

  Salesforce gives you more ways to automate than ever. Choosing the wrong one is expensive. Here is a practical framework for picking the right tool.

  FlowApexAgentforce
• 
  AI Agents Naresh Jun 10, 2026 14 min read

  AI agent security: a practical guide to prompt injection and least privilege

  Every AI agent that reads untrusted content can be turned against you, and no filter catches every prompt injection. Here's the agentic risk map, and why least-privilege architecture, not filtering, is the defense that holds.

  SecurityPrompt injectionAI governance
• 
  Platform Naresh May 26, 2026 12 min read

  Will AI replace Salesforce admins? An honest read on the 2026 job market

  Salesforce cut 4,000 support roles while its AI product line more than doubled, and admin job supply is outpacing demand. Here's what the data says is shrinking, what's growing, and how to position yourself either way.

  CareersAISalesforce admins
• 
  Agentforce Naresh May 19, 2026 11 min read

  Agent Script: when deterministic control beats prompt engineering in Agentforce

  Prompt-tuned agent behaviour is probabilistic, and regulated workflows can't run on probably. Here's what Agent Script is, how its hybrid reasoning model works, what it changes about testing, and where natural-language instructions still win.

  Agent ScriptAgentforceTesting
• 
  Data 360 Naresh May 12, 2026 14 min read

  How to cut your Salesforce Data 360 credit bill: an optimization playbook

  Most Data 360 bill shock traces to a handful of settings: streams that didn't need streaming, hourly refreshes nobody asked for, unification re-runs. Here's an eight-lever triage checklist your team can run this quarter.

  Data 360CreditsCost control
• 
  AI Agents Naresh May 7, 2026 12 min read

  Why AI agent projects fail: a field autopsy of the cancellation statistics

  Gartner expects over 40% of agentic AI projects to be canceled by the end of 2027, and MIT found most pilots return nothing. Here's what kills these projects, and the pre-launch checks that predict survival.

  AI agentsROIGovernance
• 
  Platform Naresh May 5, 2026 12 min read

  Salesforce data quality: a practical guide to getting AI-ready

  Every AI feature in your org, Agentforce grounding, Einstein scoring, Data 360 unification, amplifies whatever data sits underneath it. Here's how to find the duplicates, stale records, and gaps before your agents do.

  Data qualityAI readinessDuplicates
• 
  Agentforce Naresh Apr 28, 2026 13 min read

  The Einstein Trust Layer: what it protects, what it doesn't, and what stays your job

  The Einstein Trust Layer gets invoked in security reviews as if it settles the question. It settles about half: here's what the pipeline verifiably does, where it stops, and which risks remain squarely yours.

  Trust LayerSecurityAI governance
• 
  Data 360 Naresh Apr 21, 2026 13 min read

  Identity resolution in Data 360: a practical guide to match rules that don't over-merge

  One loose match rule can fuse two real customers into a single unified profile, and every downstream activation inherits the error. Here's how Data 360 match and reconciliation rules work, and a conservative-first sequence for tuning them safely.

  Identity resolutionData 360Data quality
• 
  Platform Naresh Apr 14, 2026 13 min read

  Salesforce sandbox strategy: a practical guide for teams of 3 to 30

  Most small Salesforce teams run either one overloaded sandbox or a sprawl of stale ones, and both end the same way: production surprises. This guide right-sizes environment topology, refresh discipline, and release cadence for teams of 3 to 30.

  DevOpsSandboxesRelease management
• 
  Agentforce Naresh Apr 7, 2026 13 min read

  The Atlas reasoning engine: a practical guide to how Agentforce decides

  Most Agentforce misfires aren't model failures, they're topic and instruction failures, baked in before the LLM ever runs. Understanding how Atlas classifies, plans, and loops turns those fields from documentation into the programming surface they really are.

  AgentforceAtlasArchitecture
• 
  AI Agents Naresh Apr 2, 2026 13 min read

  What is MCP? The Model Context Protocol explained for enterprise architects

  Every agent platform wants its own connector to every system you run, and that matrix grows multiplicatively. Here's what the Model Context Protocol standardizes, where its security risks live, and what Salesforce's MCP support means for your stack.

  MCPIntegrationStandards
• 
  Data 360 Naresh Mar 31, 2026 13 min read

  Salesforce Data 360 vs. Snowflake: the wrong question, and how to use both

  Teams keep framing Salesforce Data 360 and Snowflake as rivals, then buying the wrong tool for the job. Here's where they overlap, where they don't, and a reference architecture that uses each for what it was built to do.

  Data 360SnowflakeArchitecture
• 
  Platform Naresh Mar 24, 2026 13 min read

  Salesforce API limits: why orgs blow the daily allowance and how to fix it

  Nothing about Salesforce API limits feels urgent until integrations start failing with REQUEST_LIMIT_EXCEEDED and the sync queue backs up. Here's how the daily allowance is computed, what consumes it, and the fixes that bring usage back under the ceiling.

  APIIntegrationLimits
• 
  Agentforce Naresh Mar 17, 2026 13 min read

  How to build your first Agentforce service agent (and what Trailhead glosses over)

  Trailhead makes an Agentforce service agent look like an afternoon project; production orgs disagree. Here is the full build sequence: enablement, agent user, topics, actions, grounding, testing, deployment, including the steps the quick starts skip.

  AgentforceService CloudHow-to
• 
  AI Agents Naresh Mar 12, 2026 13 min read

  Enterprise AI agent use cases: a practical guide to what survives production

  Gartner expects over 40% of agentic AI projects to be cancelled by 2027, yet a handful of use cases keep clearing production quietly. Here's what separates the survivors from the pilot-ware, and five criteria to score your own.

  AI agentsUse casesEnterprise
• 
  Data 360 Naresh Mar 10, 2026 13 min read

  Zero copy in Salesforce Data 360: a practical guide to what moves and what it costs

  Zero copy promises live access to warehouse data with no pipelines, but acceleration, indexes, and per-row billing complicate the story. Here's what stays federated, what still materializes, and how to decide per data source.

  Zero copyData 360Architecture
• 
  Platform Naresh Mar 3, 2026 12 min read

  How to run a Salesforce security review: beyond the Health Check score

  A 95% Health Check score doesn't mean your org is secure, the tool never looks at who can see what. Here's the quarterly review we run instead, from permission sprawl to over-privileged integration users.

  SecurityHealth CheckPermissions
• 
  Agentforce Naresh Feb 24, 2026 13 min read

  Agentforce vs. Microsoft Copilot Studio: a stack-honest comparison for enterprise buyers

  Salesforce and Microsoft are selling the same dream: governed digital labour at enterprise scale. Here's where each platform wins: grounding, agent design, pricing, governance, and the one question that settles most evaluations.

  AgentforceCopilot StudioComparison
• 
  AI Agents Naresh Feb 19, 2026 12 min read

  AI agents vs. chatbots vs. RPA: which class of automation does the job need?

  Every automation pitch now says 'agent', but most tasks are cheaper and safer with a chatbot, an RPA bot, or a copilot. Here's a decision table that maps task characteristics to the class that fits.

  AI agentsChatbotsRPA
• 
  Data 360 Naresh Feb 17, 2026 13 min read

  Salesforce Data 360 pricing: a practical guide to credits, consumption, and forecasting

  Data 360 bills by consumption, and the spread between its cheapest and priciest operations is four orders of magnitude. Here's what a credit is, where they go, why streaming multiplies costs, and how to forecast a bill you can defend.

  Data 360PricingCredits
• 
  Platform Naresh Feb 10, 2026 13 min read

  Salesforce DevOps Center vs. change sets: a maturity ladder for release management

  Most painful Salesforce releases trace back to a deployment process the team outgrew years ago. This guide maps the maturity ladder: change sets, free DevOps Center, paid platforms, full CI/CD, and shows where your team should sit.

  DevOpsRelease managementCI/CD
• 
  Agentforce Naresh Feb 3, 2026 13 min read

  Agentforce pricing explained: Flex Credits, conversations, and the real cost math

  Two Agentforce quotes for the same agent can differ by a factor of four depending on which meter you pick. Here's how conversations and Flex Credits price out, what the invoice leaves off, and how to build a year-one number you can defend.

  AgentforcePricingFlex Credits
• 
  AI Agents Naresh Jan 27, 2026 13 min read

  Agentic AI vs. generative AI: the difference that changes your architecture

  Generative AI produces content on request; agentic AI takes actions to reach goals, and that distinction rewrites your architecture, risk, governance, and cost model. Here's how to tell the difference before you build the wrong system.

  Agentic AIGenerative AIComparison
• 
  Platform Naresh Jan 20, 2026 14 min read

  Salesforce integration patterns: a practical guide to what survives production

  Most Salesforce integrations work on demo day and fail eighteen months later, usually at 2 a.m. Here's how the platform's own integration patterns map to production reality, and the four failure modes behind the expensive incidents.

  IntegrationArchitectureAPI
• 
  AI Agents Naresh Jan 15, 2026 11 min read

  What are AI agents? A definition that survives contact with production

  Vendors call almost everything an AI agent, and most of those claims fall apart in production. This is the definition I use when scoping projects, the loop behind it, the autonomy dial that makes an agent governable, and a five-question test to run before you build or buy.

  Agentic AIAI agents
• 
  Data 360 Naresh Jan 13, 2026 11 min read

  What is Salesforce Data 360? The platform formerly known as Data Cloud, explained

  Salesforce's customer data platform is on its sixth name and most buyers have lost the plot. This is what Data 360 does, stage by stage: ingest, model, unify, activate, and ground agents. Plus what the rename changed, what zero copy means, and who needs it.

  Data 360Data CloudCDP
• 
  Agentforce Naresh Jan 8, 2026 11 min read

  What is Agentforce? A plain-English explainer from implementers

  Every Salesforce roadmap now has Agentforce written on it somewhere, usually without a shared definition. This is what the platform is: agents, topics, actions, the Atlas reasoning engine, grounding through Data 360, what ships out of the box, and what running one responsibly takes.

  AgentforceAI agents
• 
  Platform Naresh Jan 6, 2026 13 min read

  Salesforce governor limits: a practical guide to designing around them

  Governor limits aren't arbitrary hurdles, they're the price of multitenancy, and hitting one usually means your design owes something. Here's why the limits exist, which ones bite in production, and the patterns that keep you clear of them.

  ApexArchitectureGovernor limits

No posts in this category yet.