AI PM Accelerator
PM Interview Toolkit
Product Learning Library
Fundamentals of Product Management
PM Communication Toolkit
Being a great AI PM is hard. This newsletter makes it easy.
One clear idea per email. That’s it.
Airbnb replaced rigid phone menus with four ML models running in real time: ASR, intent detection, semantic retrieval, and paraphrasing. WER dropped from 33% to 10%.
Not every request needs your most expensive model. LLM routing sends simple tasks to cheap models and hard ones to smart ones, cutting costs 30-70%.
Netflix's recommendation engine drives 80% of viewing. Here's how they built a post-training framework to adapt LLMs for their catalogue from SFT to RL.
I built a 7-agent AI pipeline that writes my newsletter (JustAnotherPM) in 45 minutes instead of 8 hours. Here’s the full system, design decisions, and lessons learned.
Pre-training makes a model smart. Post-training makes it useful. Here's how SFT, RLHF, DPO, and Constitutional AI work, explained for PMs.
Prompt caching doesn't cache the answer. It caches the question. Here's how it works, what it costs, and how teams are cutting down their LLM bills.
Evals are replacing PRDs as the primary spec for AI products. Here's why — and how to cover single-turn, multi-turn, and agent evals in 2026.
Anthropic's agent containers kept dying. So they borrowed from OS design and cut latency by 90%. Inside the Managed Agents architecture.
Stripe and OpenAI co-built ACP, an open standard for AI agents to make purchases. It already powers Instant Checkout in ChatGPT. Here's how it works.
Claude Channels lets external systems push events into a live Claude Code session. Here's what it is, how it works, and why it matters for AI product teams.
Classic RAG retrieves once and generates. Agentic RAG retrieves in loops, checks what it found, and goes back for more. Here's what that means for the AI products.
Anthropic now has three products called Claude. One is a chatbot, one writes your code, and one manages your files. Here's how they're different and when to use each.
