Anthropic’s first technical PM on token maxing, the jagged edge, and living in the future | Dianne Penn
Summary
Lenny Rachitsky interviews Diane Penn, Head of Product for AI Research and Labs at Anthropic, discussing the company's evolution from a small startup to an industry leader. They explore key inflection points like Opus 3 and Claude Code, emphasizing that "evals are the new PRDs" in AI product management. Penn highlights the importance of adaptability, first-principles thinking, and ambitious experimentation as models rapidly advance.
Topics discussed
Introduction: Diane's role at Anthropic and early skepticism
Early days at Anthropic: Culture, small team, and first models
Sponsor Break: WorkOS
The inflection point: Opus 4.5 and the rise of Claude Code
Inside the AI lab: Evals as the new PRDs and measuring progress
Token maxing and living in the future to understand user value
Anthropic Labs: Discontinuous bets and experimental culture
Product management for research: Actionable feedback loops
Career advice for AI PMs: Ambition and first-principles thinking
Safety, Mythos, and the evolving pre-release process
Sponsor Break: Mercury
Hiring AI PMs: Traits, evals, and the death of traditional PRDs
Finding joy in AI: Experimentation and avoiding burnout
AI as a thinking partner: Augmenting human judgment and writing
The future of human value: Inner voice and curiosity
Work-life balance: Team support and the 'mind meld' culture
Lightning Round: Personal insights, movies, and career lessons
Listen ad-free on Castria