COMPASS logo

COMPASS Research Group

COoperative Machine intelligence for People-Aligned Safe Systems

The COMPASS group at the ELLIS Institute Tübingen and the Max Planck Institute for Intelligent Systems investigates how to build AI systems that are safe, aligned with human values, and robust against adversarial manipulation. We work on broad topics in A(G)I safety and security: interpretability, reasoning, evals, contextual integrity, agentic risks and opportunities, multi-agent dynamics, agents with long-term memory, self-improving agents, (deceptive) alignment, situational awareness, manipulation, and deception.

Members

Join us

Join COMPASS!

I'm looking for curious, driven researchers interested in AI safety and security. We are based in Tübingen, Germany. No knowledge of German is required, and for PhD students, a publication record is not a prerequisite: any relevant experience, passion, curiosity, research potential, and willingness to learn matter most.

Topics we're currently hiring for

Open calls for specific projects

General positions

What I'm looking for

How to apply

  1. Fill in the interest form (button above) for any of the general positions, with a statement as described above. For an open call, use the form linked in that call instead.
  2. PhD applicants can additionally apply through one of the graduate programs: ELLIS PhD Program, Max Planck & ETH Center for Learning Systems (CLS), or IMPRS-IS.
  3. I review each application and reach out if there's a good fit.

FAQ

Alumni