Product philosophy
A living list of beliefs. Updated whenever a new conviction solidifies.
- 1
"AI Product" is a term I feel ambivalent about. AI is a technology — like Java. The real job is to solve a problem. AI is a means to the end, not the end itself.
- 2
Understanding the limitations of models, the unglamorous data labeling work, and driving complements from engineering and UX work — that's the difference between shipping good and great AI product.
- 3
Product sense is knowing what users need most out of all possible requests (and which makes the biggest business impact). Between what users ask for and what they actually need — that gap is where product work lives.
- 4
Be a multiplier — PMs are measured by the team's output. The best PMs make engineers more effective, designers braver, data scientists more product-minded.