11+ years building SaaS and AI-powered products across hospitality, travel, and healthcare, most recently leading GPT-4-based classification at Expedia and AI-powered inventory ranking at Hilton Grand Vacations. Lately I've been pairing that product judgment with hands-on prototyping: using Claude, Cursor, and Lovable to go from idea to working demo before a single line of a spec gets written.
A small Claude-powered agent that turns a rough feature idea into a structured PRD: problem statement, user stories, and acceptance criteria. It decides on its own when to pull in market context, flag compliance risk, or format the final document.
In active development. Demo and write-up coming soon.
Replace with your own project. Example shape: a retrieval-augmented assistant over a set of documents, with an eval harness scoring answers for accuracy and groundedness against a held-out test set.
Replace with your own project. Example shape: a script that runs one prompt set against several models and scores them on accuracy, latency, and cost: the kind of build-vs-buy evidence that's hard to fake in an interview.
