Postdoctoral researcher positions focused on the use of advanced AI techniques such as reinforcement learning and large language models for complex system optimi-zation

Track A – RL / Optimization

• Design, implement and evaluate RL frameworks for complex, high‑dimensional environments, leveraging simulation‑based optimization and digital‑twin testbeds
• Explore multi‑agent or distributed training techniques to scale optimization across interacting subsystems
• Collaborate with domain scientists to translate algorithms into prototype decision‑support tools

Track B – LLM / Knowledge‑Engineering

• Develop and fine‑tune LLM pipelines that integrate structured reasoning tools (e.g. retrieval‑augmented generation, knowledge graphs, constraint parsers)
• Build workflows for transparent explanation and evaluation of model decisions
• Maintain and contribute to open‑source libraries that support reproducible research in the project

Must have (both tracks)

• PhD (or equivalent) in computer science, applied mathematics, operations research, physics, statistics or a related field
• Documented, hands‑on track record of applying RL and/or LLMs to real‑world problems (publications, deployed systems, or open‑source artefacts)
• Excellent Python skills and familiarity with modern ML stacks (e.g. PyTorch, JAX, Hugging Face)
• Ability to thrive in an interdisciplinary environment and to communicate complex ideas clearly
• Prior experience in healthcare, epidemiology, or medical applications of machine learning or statistics (e.g., causal inference) is not required

Nice to‑have – Track A

• Experience with simulation‑based optimization or digital‑twin frameworks
• Familiarity with multi‑agent RL or distributed training

Nice to have – Track B

• Experience integrating LLMs with structured reasoning tools (RAG, KGs, constraint solvers)
• Track record of open‑source contributions to RL, LLM or optimization libraries