AI Product Engineer; m​/w​/d) @Munich

You take over our platform. Not a ticket queue, not a slice of it — the whole thing. Product surface, data layer, infrastructure and the agentic workflow we build with. You work alongside our Technical Lead and own how it gets built.

Full-time, office-first in Munich. Working from abroad is possible in exceptional cases, with regular travel to Munich required and fully covered. Fluent English is a must, German is not.

Aybee shows how people really buy. Our AI software recreates digital shopping environments — Amazon, D2C shops, shelves - where real consumers decide about products, packaging, prices and ads under realistic conditions. We turn the data we collect into tailored, AI-powered insights that help our clients make confident decisions.

Our clients include brands like The Quality Group (ESN, MORE NUTRITION), BLACKROLL, Bayer, EMMA, Barilla and asambeauty. They run their launches, pricing calls and packaging decisions through the tests we build.

We bring the requirements. You decide how they get built. You will:

• Own the product surface: Next.js App Router, Type Script, React, Tailwind, shadcn/ui — from the test setup our CS team uses daily to the reports that land on a CMO's desk.
• Own the data layer: Supabase (Postgres, Auth, Edge Functions, RLS), including the statistical pipeline that turns raw session data into significance-tested results clients act on.
• Own the infrastructure: Cloud Run, Git Hub Actions, staging/production separation, versioned migrations. If it breaks, you're the one who fixes it.
• Architect the solutions: You take a requirement and come back with how it should be built, what it costs us, and what we're trading away. Then you make the call and stand behind it.
• Maintain and extend our agentic workflow: Our orchestration and quality gates are already in place and central to how we ship. Keeping them sharp is part of the job.
• Ship AI features with the team: You work with our Technical Lead and our AI engineer to bring new AI capabilities into the product — they build the AI side, you own how it lands in the platform.

• Extreme ownership: You've been the engineer in the lead — the one accountable for what ships and on call when it doesn't. Nothing stays half-finished on your watch.
• Architectural judgement: You're comfortable making consequential calls on trade-offs — performance against complexity, speed against maintainability — and explaining the reasoning to people who aren't engineers.
• Command of the stack: Next.js, Type Script, React and Postgres/SQL.
• Real agentic experience: You've spent at least a year using Claude Code, Cursor or agent orchestration to ship production software — not just experimenting on weekends. You know how to use these tools to move faster without creating technical debt.
• 3+ years shipping to production in a fast-paced environment, with full ownership of what you shipped.
• Fire: You push back when a plan doesn't hold up technically, decide without being asked, and want this platform to be yours for years rather than months.
• High-pace mentality: We work together in the office, at speed and with high standards. Our team works hard, and when it matters, longer.
• Statistics or experimentation background
• CS degree or equivalent
• German

Five steps, and we move fast through all of them:

• Intro call with our Technical Lead. What you've built, what you're looking for, and whether this is the kind of ownership you actually want.
• Intro call with our CEO. The business, where we're going, and your questions about both.
• Case study call with our Technical Lead and AI engineer. A real problem from our platform. We want to see how you think through the trade-offs, not whether you can recite algorithms.
• Final round with our…