Retrosynthesis Researcher, Machine Learning
Core
Develop and implement AI/ML models for retrosynthetic pathway prediction, reaction outcome prediction, and novel synthetic route identification to advance small molecule drug discovery and materials science.
Role type
Research-level machine learning engineer (computational chemistry)
Builds
Production ML models for retrosynthesis and reaction prediction integrated with cheminformatics platforms
Domain
Computational chemistry / AI for drug discovery and materials science
Deliverable
production ML models
Required skills
Graph neural networks, transformer-based models, deep learning for reaction prediction, cheminformatics (RDKit, Open Babel), Python, PyTorch, TensorFlow, JAX, organic synthesis and reaction mechanisms
Preferred skills
Chemical reaction databases (Reaxys, USPTO, Pistachio), CASP tools (AiZynthFinder, ASKCOS, IBM RXN), graph-based learning, attention mechanisms, reaction condition prediction, Schrödinger Suite, de novo design, generative ML, cloud/HPC, quantum chemistry (DFT)
Technologies
PyTorch, TensorFlow, JAX, RDKit, Open Babel, Schrödinger Suite, LiveDesign
Responsibilities
Develop AI/ML models for retrosynthetic pathway prediction; Apply deep learning to predict reaction outcomes and optimize conditions; Curate and manage reaction datasets from literature and patents; Integrate retrosynthesis tools with cheminformatics platforms; Collaborate with synthetic chemists to validate predicted routes; Contribute to scholarly publications and represent the research group at conferences
Seniority
Senior, hands-on IC researcher