About this wiki
The AI Product Engineer Wiki is a free guide to becoming an engineer who owns AI features end to end: from the user problem and the experience, through prototypes and evals, to launch, metrics and cost.
The author
Ghassen Khabacha is a senior full-stack engineer building AI products with Node.js, React and Next.js, and also wrote the sister site, the AI Full-Stack Engineer Wiki (opens in a new tab), which covers the engineering side of building with LLMs, from web foundations to multi-agent systems.
This wiki focuses on the product side: deciding what to build, designing AI experiences people trust, measuring quality with evals, and shipping and pricing AI features responsibly.
What it covers
51 articles across 9 stages, 6 portfolio projects and 7 reference pages. The stages, in order:
- Start Here: Understand the role, the AI product lifecycle and pick a learning plan that fits your week.
- Product Foundations: Find problems worth solving, write a one-page spec and decide whether AI is the right tool at all.
- LLM Capabilities for Product Decisions: Turn model strengths, latency, cost and context limits into product decisions you can defend.
- Designing AI UX: Design interfaces that set expectations, show their sources and keep people in control.
- Rapid Prototyping: Go from idea to a tested demo in days, then turn what you learned into a spec.
- Evals as Product Metrics: Measure quality the way users experience it and catch regressions before they ship.
- Shipping & Iterating: Roll out gradually, instrument everything and turn user feedback into measurable improvements.
- Trust, Safety & Economics: Defend against misuse, protect user data and make every AI feature pay for itself.
- Projects & Career: Turn your projects into a portfolio and case studies, and prepare for AI product interviews.
Every article includes diagrams with text descriptions, a "Product lens" on the decision it informs, common mistakes, a hands-on exercise and a short quiz. Code examples use TypeScript and the Anthropic SDK, and are type-checked against the real SDK on every change.
How it's maintained
Last updated . Model names and prices live in one configuration file and are checked against the provider's documentation; check the model cheat sheet for the date they were last verified. Statistics are cited or clearly marked as opinion. If you find an error, please get in touch.
License
- Code examples are licensed under the MIT license (opens in a new tab): use them freely in your own projects.
- Written content is licensed under CC BY 4.0 (opens in a new tab): share and adapt it with attribution.