Agents

What I'll be writing about

Notes on building with AI agents, and the parts the demos skip.

· 1 min read

TL;DR · 30 seconds

  • This blog is about building with AI agents after the demo works.
  • Expect posts on observability, context, evals, MCP and guardrails.
  • Every post starts with a TL;DR like this one.

Why another AI blog

Most writing about agents is about demos. I want to write about what comes next: making agent systems you can see, afford, trust and hand to other people.

What to expect

  • Observability. You can’t improve a multi-agent system you can’t see.
  • Context engineering. What goes into the window, and what it costs.
  • Evals and guardrails. Deterministic checks for non-deterministic systems.
  • MCP in practice. Putting internal tools inside the chat clients people already use.

How posts are built

  1. Real problem
  2. What I tried
  3. What broke
  4. What worked

If you’re working on similar problems, get in touch by email or on LinkedIn.