Become anAI Product Engineer

You don’t need to know how to code. We start from the fundamentals and give you the bases —plus two books, one on software and one on AI— so you understand everything from scratch, with simple questions and no jargon. That said: if you’ve ever touched SQL or written something in Python or another language, it’ll be easier, and not being afraid to read code helps. But it’s never strictly required: with the progress of LLMs it matters less and less, and understanding the architecture concepts is enough.

You can take it, with one caveat: this path is focused on building AI Fluency and Tech Fluency — not on teaching you product. We don't explain product management or engagement concepts. If you already understand the classic product concepts (whether you come from business or elsewhere), you'll get a lot out of it. In fact, we see people from operations or performance marketing get a huge amount out of it. If what you're after is Product Fluency, we recommend a personalized program inside our platform.

It's not mandatory, but we recommend around 3 years of experience in a squad: that constant exposure means many concepts you'll see here will already feel familiar. It's easier if you've worked side by side with engineers — even in another role — because you'll have been exposed to key concepts (for example, development vs. production environments). Without that, you can still follow it, but it'll take a bit more effort.

More than you’d think, and at a higher level. Coming from Product Manager, you level up to AI Product Manager; coming from design, you become a Design Engineer. And if you really want to build, a Product Builder, a Maker or —as we like to call it— a Product Engineer, even an AI Product Engineer. New roles open up too: AI Consultant, AI Coach… The common thread: you stop depending on a technical team to make your ideas real, and that multiplies your value in any role.

Yes, it's separate. Claude Code is Anthropic's and is billed apart from the course. We recommend the Max plan (around 90€/month): if you go all in, a single month is enough to complete the full path and the 4 courses. We show you how to set it up from the very start.

We recommend Claude Code — it's what the course is built around and what gets you the productivity jumps. OpenCode or Codex can also work. If you prefer Cursor or Antigravity, you can try, but unless you use a console (CLI) harness like Claude Code or OpenCode, the productivity gains will be smaller. We don't recommend GitHub Copilot: you won't reach the productivity gains or be able to follow the course properly.

Both programs include the core of the path: the 4 courses, the 2 books, and all the asynchronous content. Self-paced is 100% on your own time, with immediate access to start today. Live adds the support: live classes and workshops, 1:1 mentoring, a boilerplate for your apps, and access to the new courses we keep launching. In both, you keep access to the recorded content for as long as the platform stays active.

It depends on your background and how many times you need to rewatch each concept, so we give a rough range based on what our students tell us: around 80 hours of dedicated work. At about 10 hours a week, you’ll have it done in roughly 2 months. And they’re hours worth putting in: you walk away with skills that stay with you for the rest of your life and at least double your productivity. If you join a Live edition, add the live sessions (classes, workshop and mentoring) on top of those ~80h at your own pace.

There's no fixed, guaranteed number. We usually open one or two Live editions a year — typically in summer and at Christmas — but it depends on how the product and the market evolve. The previous one ran in December 2025 and the latest in July 2026; there may be another in December 2026, though it's not confirmed. If there's an edition open right now, don't let it pass — the next one may take a while. And if there's no Live planned, former students always get priority, with very, very special discounts based on how many courses you've taken with us.

Yes. You can split the payment into installments with Klarna. Just sign up and, when you buy the path or any course, pick Klarna as your payment method at checkout. One thing to keep in mind: offers and promotions don't apply when paying with Klarna — the installment plan is calculated on the standard price.

We start from zero and take you from 0 to 100 with everything you need, designed for Product profiles — where you’ll get the most out of it. We teach the fundamentals of AI from real experience: 4+ years working in AI (since neural networks) and the last two with LLMs, building AI SaaS — with the real mistakes that leaves behind. And we teach the bases that don’t expire —for example, the difference between a skill and an agent—: that’s already there and doesn’t change even when the tool does. On top of that, we’re in the process of getting certified as Claude Architects and becoming an Anthropic partner.

There are 4 because the path is layered: each course levels you up, from the fundamentals to agentic systems in production. Each course is a set of sessions, and each session teaches you one skill end-to-end. That set of e2e skills is exactly what lets you finish building something complete, on your own and at a professional level. Inside, each session has several microconcepts —short 5-to-15-minute videos, one concept each— and two Trainings of around 20-30 minutes where you apply it; after each video, a question to lock in what you learned. All at your own pace and there to rewatch whenever: move through the concepts on the train, the bus or walking —listening or watching from your phone— and when you get half an hour, sit down and do the training.

Yes. We work with teams so they go from using AI to building with judgment. Write to us and we'll tailor it.

We're registered in Anthropic's Claude Partner Network and on our way to becoming a Partner. The whole path is built around Claude Code, the tool we work with every day.

The “Founding” shapes how we teach. A Founding Engineer is usually the first person to join a project: they come in when there’s no team yet and have to stand up almost everything. That’s why we explain every single thing with that founder’s mindset —the why behind each decision— so you can build it on your own and have it scale without needing anyone else. A Product Engineer (or Design Engineer) builds product by combining product/design with engineering, usually within a team and on something that already exists. The Founding version is that same profile, trained to start from zero with a founder’s judgment and full autonomy. And that’s where the AI Product Engineer comes in: with the right knowledge and Claude Code —or any agentic development harness—, a single person does what used to take a squad of 10. That’s exactly what comes out of the path.