4 sessions on AI agents at APE XXL 2026 — 8 December at Dierenpark Apenheul.
copilot-instructions.md, AGENTS.md, SKILL.md, prompt files — they all tell the agent how to behave. So why does one get followed immediately while another disappears into the void? A practical mental model for how AI instruction files are actually loaded and applied — and how to spot guidance that is being ignored versus never loaded.
A build-and-break story from production: 32 specialized AI agents assess a live Azure estate against the Well-Architected Framework — and the hardest part was never the prompt. Watch the system produce a plausible finding that is wrong, then catch its own hallucination live. Three reusable patterns for agent output you can actually trust.
AI agent memory sounds complicated until you realize there is already a perfect example of it: your own brain. Each type and technique of agent memory matched to the part of the brain it mirrors — then shown at work in a real agentic system in production: how it decides what to remember, how it recalls the right thing at the right moment, and how it forgets the rest.
Building one AI agent is easy; engineering a platform for thousands is another story. Which department is burning the most tokens? Does every agent in production follow your safety guidelines? A reference architecture for a centrally governed AI platform on Azure — token rate limiting, semantic caching and security through API Management and Microsoft Foundry, agent identity in Entra ID, deployed as a hub-and-spoke landing zone in code.