Writing

Notes on building AI systems that have to work in production — mostly the unglamorous parts. Written from the perspective of someone who spent 15 years on enterprise backends before the current wave.

· 14 min read

Your Agent Loop Is a State Machine, Not a While Loop

Everyone's first agent is twenty lines: call the model, run the tool, append the result, repeat. What has to go in between those steps — validation, tool scoping, approvals, bounded execution, re-validation — before it survives real users.

Read →
· 9 min read

What Telecom Monitoring Taught Me About Debugging LLM Agents

Before I worked on agents, I spent a year building business activity monitoring for a national telecom. Almost everything useful I know about debugging LLM systems, I learned there — and the industry is currently relearning it the hard way.

Read →