SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
commercetools is seeking an AI Engineer to own AI capabilities end-to-end, from data and context that feed models through agent orchestration to the generative UI users interact with. This is a full-stack role where you will personally ship and operate production applications across backend, frontend, data, and ML—not coordinate across teams.
You will build agent loops and orchestration systems with tool use, structured outputs, business-rule enforcement, and graceful failure handling. You'll construct the integration layer (via MCP) that enables agents to securely call backend systems and business APIs. You'll render model output into live, streamed, accessible UI via generative-UI protocols and ground models in real data through retrieval, embeddings, and feedback loops that actually close.
You'll treat evals as a first-class system focused on provable improvement rather than intuition, and take capabilities from prompt-in-a-notebook to measured, guard-railed production features. This role is hybrid and available in Berlin, Munich, London, or Valencia.
You are a creative problem-solver wired to find solutions, confidently diving into complex challenges and making them simple for others. Your curiosity drives constant growth in an environment of trust and teamwork. Your unique perspective matters more than checking every box.
REQUIREMENTS:
- Shipped full-stack apps to production and owned them end-to-end: frontend through backend to deploy and operate, not handed off at a layer boundary
- 5+ years shipping production software; 2+ years building with LLMs in production (real users, not demos)
- Hands-on experience with UI protocols (A2UI, AG-UI, MCP-UI, or similar); you've built against a generative/agent-driven UI protocol, not just consumed a REST API
- The AI spine: prompt/context engineering, agents, tool use, structured outputs, evals, MCP Python and TypeScript/Node + React; comfortable with async, streaming, and observability
- Working fluency with retrieval/embeddings and SQL
NICE TO HAVE:
- Fine-tuning/distillation experience
- E-commerce or marketplace systems knowledge
- Track record in 0→1 product development
