Founding RL Researcher (Remote, London/Paris preferred)

About the role

We are building a pre‑seed AI startup focused on long‑horizon reinforcement learning using real expert behavioural data. As a founding RL researcher you will shape the scientific agenda from day one and work closely with the engineering team to turn research into production‑grade infrastructure.

Key responsibilities

• Define the research roadmap for long‑horizon RL on real expert data.
• Design novel reward models, verifiers, and evaluation frameworks where no textbook exists.
• Run training experiments that demonstrate genuine capability gains.
• Publish research and influence how the field thinks about environment design.
• Collaborate with engineers to build production‑ready RL environments.

Required profile

• Proven experience in frontier RL research (lab, research‑heavy startup, or strong PhD with publications).
• Passion for environment design as a core AI problem.
• Interest in long‑horizon RL over short‑form RLHF.
• Desire to define scientific direction rather than follow an existing stack.
• Fast‑moving, first‑principles thinker who wants ownership of foundational decisions.

Required skills

What we offer

• £300,000 base salary plus founding equity at a ~$30M pre‑seed valuation.
• Frontier AI labs as paying clients from day one.
• Hybrid work model with London/Paris preference and remote flexibility.
• Opportunity to shape a brand‑new research agenda in a high‑growth startup.