Setting an ALM strategy before the agents pile up
As agents scale across an org, ALM gets confusing fast. Use this process - starting with cost centres to plan your environment strategy.
Recent Posts
The agent isn't your orchestrator
Why Copilot Studio agents struggle to loop through hundreds of questions, and how a Power Automate fan-out pattern fixes it without rebuilding the agent.
Why your agent's knowledge breaks on import (and how to fix it)
When you promote a Copilot Studio agent through ALM, the Dataverse search connector behind your knowledge source doesn't recreate itself. Here's the fix.
Meetings are prompts now
Meetings aren't about efficiency anymore. They're the richest context you'll ever give an AI. The question is what happens after.
The long way round to a knowledge-grounded HR agent
How building a custom chunking pipeline for an HR triage agent taught me everything Microsoft was about to solve natively.
Short-Term Memory in Copilot Studio: Extending the Context Window
Short-term memory in Copilot Studio isn’t about persistence - it’s about continuity. By promoting confirmed conversational state into session memory between execution passes, agents maintain context, avoid drift, and behave consistently across multi-step interactions.
From Rigid Flows to Adaptive Agent Design
Most Copilot Studio agents fail by being too rigid. Clear roles matter, but over-prescribed flows turn agents into brittle utilities. Flexible execution—planning, checking, adapting—is what makes agents useful beyond demos.
Structuring Topics in Copilot Studio for Agentic Behaviour
Explore how to build Copilot Studio agents that reason and plan through multi-step actions. Learn to structure topics with inputs and outputs for agentic behaviour, non-deterministic outcomes, and richer, context-driven responses.
Grounding Copilot Studio Prompts on Dataverse Tables
This article explains how Copilot Studio grounds custom AI prompts on Dataverse tables, enabling precise, structured, and filtered data retrieval.