Applied AI Scientist, Simulation & Reinforcement Learning

Skills

About the Role

You will own and develop a SimPy-based discrete-event simulator, build experiment infrastructure, and reduce the sim-to-real gap using live site data. You will turn the simulator into an RL environment, design, train and evaluate RL policies, and communicate results to stakeholders.

Requirements

• PhD in a quantitative field or MSc with 5+ years of industry experience applying AI/ML or simulation
• Hands-on experience building or validating simulators, ideally discrete-event (SimPy)
• Practical reinforcement learning experience with Gym-style environments
• Strong Python and software engineering fundamentals
• Rigorous experimental and statistical thinking
• Experience in robotics, logistics, warehouse or operations-heavy domains
• Publications in major AI, ML or robotics venues
• Excellent English communication skills
• Must reside in the UK with legal right to work in the UK

Responsibilities

• Own and develop the SimPy-based discrete-event simulator
• Pay down technical debt and improve test coverage, code quality and tooling
• Build experiment infrastructure for orchestration, study management and automated runs
• Generate and maintain site data needed to simulate customer warehouses
• Support scenario testing and sensitivity analysis
• Define, measure and reduce the sim-to-real gap
• Run parameter optimisation and tuning studies
• Model warehouse processes such as congestion, picker behaviour and queueing
• Communicate results and confidence to stakeholders
• Build Gymnasium-style RL environments
• Design, train and evaluate RL policies against heuristic baselines
• Prototype and productionise learning-based components
• Stay current with RL, multi-agent systems and simulation research
• Contribute to technical papers, patents and publications

Benefits

• Bonus
• Equity
• Remote position