Field notes from the studio
Practical write-ups on shipping AI agents, RAG systems, and data infrastructure into production. No fluff, just what worked.
Your Agent Traces Are Lying to You: The Race to Standardize Agent Observability
OpenTelemetry just shipped semantic conventions for agent spans, Datadog adopted them natively, and the Linux Foundation stood up a new governance body for agent infrastructure. Here's what's actually usable today, what to skip, and why your traces still might not tell you why the agent did what it did.
Read articleAll articles
15 posts
Building RAG systems that answer accurately instead of hallucinating
Retrieval quality, not the model, decides whether your RAG system is trustworthy. A practical look at the retrieval choices that keep answers grounded, and the evals that keep them that way.
Read more
What it actually takes to ship an AI agent to production
The gap between an agent demo and a production agent is enormous. Here are the parts nobody shows in the demo: tight tools, guardrails, observability, and a human in the loop.
Read more
Integrating Claude into an existing product without a rewrite
You do not need to rebuild your app to add real AI. A pragmatic pattern for dropping Claude into a production codebase with evals and guardrails from day one.
Read more