Andrew Lee
AI Product Manager & Strategist. Turning commercial goals into reliable, measurable AI systems.
I work on applied LLM systems: taking models from a promising demo to something a team can ship, measure, and keep improving. Most of what I write here comes from projects I have actually built, with the numbers and the dead ends included.
Right now I'm focused on evaluation, agent reliability, and the product decisions that decide whether an AI feature survives contact with real users.
Latest
- [Insight]Hello world: how this site is built
A test post that exercises headings, code, images, links and tags. Delete once the first real post is live.
- [Experiment]Sample lab entry: what a Lab item looks like
A placeholder showing the category badge, demo and repo links. Replace with your first real experiment or case study.