Read the original at Lenny’s Newsletter ↗lennysnewsletter.com · subscriber post
By Lenny Rachitsky · lennysnewsletter.com · @lennysan on X · YouTube · LinkedIn
Amir Klein, writing via Lenny's newsletter, explains how he built an AI 'second brain' using ChatGPT Projects while launching monday.com's first AI agent. He covers giving the project a defined personality, feeding it fragmented context as text files, and using it for tasks like sign-up forms, prototype prompts, and tailored communications. His core argument is that offloading context amplifies a PM's reasoning rather than replacing it.
Subscriber post — summary only01Key takeaways
- Store scattered product context in an AI project so you can reason about it instead of holding it all in memory.
- Give your AI project a defined personality that challenges weak spots and pushes back on ideas.
- Treat nearly every work artifact, including decks, Slack exports, docs, and CSVs, as text you can upload.
- Use the same context base for many tasks, from forms and prototypes to audience-specific updates.
- The PM still makes the decisions; the second brain removes the friction between knowledge and output.
“AI won’t replace you, but a person using AI better than you might.”Unnamed (quoted by Lenny Rachitsky) · Lenny’s Newsletter
“You don’t just remember more. You reason better.”Amir Klein · Lenny’s Newsletter
Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.