Build realistic environments for AI agents and the infrastructure for RL training and benchmarking.

Location: Tübingen / Hybrid

Employment: Full-time

The role

You will build the environments and benchmarking systems at the core of our products. The role combines strong software engineering with practical work on AI agents: designing realistic environments, investigating agent behavior, and making evaluations reliable and useful. You will work across these problems with the rest of the team, with a particular focus on software architecture, infrastructure, and implementation.

What you’ll work on

Build and maintain production software systems and realistic environments for AI agents

Develop reliable backend and cloud infrastructure for RL training and benchmarking

Turn research prototypes into dependable tools and products

Improve developer workflows across backend, data, and machine-learning systems

What we’re looking for

Strong software-engineering fundamentals and a record of shipping systems end to end

Experience with backend development, infrastructure, or distributed systems

Good judgment in ambiguous projects and the ability to choose practical designs

Interest in AI systems and working across engineering and research

Useful experience

Early-stage startup or small-team product development

LLM APIs, agent systems, or ML platforms

Cloud infrastructure, containers, queues, or distributed workloads

Developer tools, evaluation platforms, or data-intensive applications

What we offer

Competitive compensation with equity

Flexible hours and a remote-friendly setup

Vacation, learning, and conference budget

Research freedom in a high-trust environment

Work with us

Submit your CV, a short note, and links to work that shows how you think and build using the application form.