Postdoctoral Researcher, AI for Retrosynthesis and Reaction Prediction

About the role

CBS is looking for a Postdoctoral Researcher to join an interdisciplinary team working on artificial intelligence-driven retrosynthesis and reaction prediction. The successful candidate will build advanced machine learning models to automate and optimise retrosynthetic analysis, helping discover efficient and sustainable synthetic routes for complex molecules in chemistry, pharmaceuticals and agrochemicals. The work sits at the intersection of computational chemistry and AI.

Key responsibilities

• Develop and implement AI/ML models (e.g. graph neural networks, transformer-based models) for retrosynthetic pathway prediction.
• Apply deep learning to predict reaction outcomes, optimise reaction conditions and identify novel synthetic routes.
• Curate and manage reaction datasets from literature, patents and proprietary sources to train and validate models.
• Integrate retrosynthesis tools with cheminformatics platforms and molecular modelling software.
• Collaborate with synthetic chemists to experimentally validate predicted routes and optimise laboratory workflows.
• Contribute to publications in high-impact journals and represent the group at conferences and workshops.
• Mentor graduate students and take part in capacity-building activities.

Required profile

• Ph.D. in Chemistry, Computational Chemistry, Cheminformatics, Computer Science or a related field.
• Proven expertise in machine learning or deep learning applied to chemical reaction prediction or retrosynthesis.
• Publication record showing expertise in AI and computational chemistry.
• Excellent problem-solving skills and ability to work in a multidisciplinary research team.

Required skills

• Python programming and familiarity with TensorFlow, PyTorch or JAX.
• Cheminformatics tools (e.g. RDKit, Open Babel) and databases (e.g. Reaxys, USPTO, Pistachio).
• Preferred: graph-based learning, transformers, CASP tools (AiZynthFinder, ASKCOS, IBM RXN), DFT and HPC experience.

To apply, send a cover letter, a CV with publications and contact details for two academic referees as a single PDF.