Practical AI: agents, automations, evals, and the production parts.
Notes on building AI systems that survive contact with real users — evaluation, guardrails, and the unglamorous parts of shipping models into production.
Why your first AI agent should be embarrassingly small
The agents that work in production tend to start tiny — one task, one human in the chair next to them, a tight feedback loop. The flashy demo can come after.
February 10, 20263 min readModel selection isn't a model decision
Picking the right LLM is more about your evaluation pipeline than about any single model's benchmarks. The model you can swap is more valuable than the model you can't.
September 28, 20253 min readBuilding a private LLM that knows your business
Off-the-shelf chatbots hallucinate when asked about your business. The fix isn't a better model — it's retrieval, the plumbing around the model.
May 16, 20254 min readAI in 2025: a year of audit, not adoption
For most mid-sized businesses, 2025 isn't going to be the year of AI adoption — it's going to be the year of AI audit. The tools have already arrived. Nobody's counted them yet.
January 1, 20253 min read