The Blog
Practical Salesforce engineering – Apex patterns, integration architecture, and platform decisions explained.
-
Grounding Agentforce responses: retrieval patterns that stop an agent making things up
Agentforce will invent a plausible answer the moment retrieval fails. Here’s how Knowledge, Data Cloud vectors and prompt grounding actually stop that.
-
Agentforce in a contact centre: where it helps the agent, and where it should stay out of the way
Agentforce is great at summarising cases and drafting replies. It has no business near refunds, angry customers, or the edge case nobody wrote down.
-
Flow orchestration for agents: handing a multi-step process to Agentforce safely
Handing Agentforce a multi-step Flow takes more than an action label. Here’s how to build input contracts, checkpoints, and approval gates that hold.
-
Designing Apex actions for Agentforce: input validation an agent can’t talk its way around
An Agentforce agent fills in your Apex action’s parameters itself. A good description won’t stop it guessing wrong. Here’s how to validate properly.
-
Zero Copy and Data Cloud: when you actually need it, and when a sync job still wins
Zero Copy sounds like it replaces every sync job in Data Cloud, but it doesn’t. Here’s exactly where it earns its place, and where a sync job still wins.
-
Agentforce and PII: what a Salesforce technical lead needs to get right before go-live
Agentforce reads PII into prompts and transcripts in ways field-level security was never built to stop. Here’s the go-live checklist that actually catches it.
-
Data Cloud ingestion patterns: getting external data into Salesforce without a mess
Batch, streaming, or native connector: the right Data Cloud ingestion pattern depends on the question you need answered, not the hype around it.
-
Topics, actions, and reasoning: debugging an Agentforce agent that gives the wrong answer
An agent that answers wrong doesn’t crash, it just sounds confident. Here’s exactly where the reasoning breaks down, and how to find it before customers do.
-
Agent observability: what to monitor once Agentforce is making decisions in production
Agentforce can act now, not just answer. Here is what Salesforce teams actually need to monitor once an agent starts making decisions in production.
-
Prompt Builder vs Apex: where to draw the line when building an Agentforce action
Prompt Builder or Apex for your next Agentforce action? Here is the decision framework for where language ends and business logic has to take over.