AI/ML Scientist

Role overview Develop machine learning systems for lab-in-the-loop molecular design and optimization in a drug discovery setting. The work spans molecular sequences, structures, conformational ensembles, and property prediction, combining research-oriented model development with production training, deployment, and data pipelines. You’ll collaborate with ML scientists, engineers, and drug designers to turn new methods into useful scientific applications.

Responsibilities - Design, train, optimize, evaluate, and deploy deep learning models, including generative and multimodal approaches. - Build data pipelines and methods for connecting omics, imaging, and clinical datasets. - Scale training and inference using multi-GPU and cloud infrastructure. - Create and maintain MLOps workflows for deployment, versioning, monitoring, robustness, and reproducibility. - Work with cross-functional teams to apply machine learning to disease research and clinical decision support.

Requirements - A bachelor’s, master’s, or doctoral degree in computer science, machine learning, computational biology, data science, statistics, mathematics, or a related quantitative field. - One to five years of relevant experience, including postdoctoral work. - Expert-level Python and strong foundations in algorithms, data structures, and software engineering. - Extensive experience with PyTorch, JAX, or TensorFlow. - Experience with distributed training approaches such as DDP, Ray, FSDP, or DeepSpeed, and deployment tools such as Triton or ONNX. - Strong debugging, problem ownership, and communication skills across technical and non-technical audiences.

Nice to have - Hands-on experience with geometric deep learning, cofolding models, or neural force fields is strongly preferred. - Background in drug discovery, genomics, or medical imaging is useful but not required. - Experience with large or small molecule datasets, multi-node GPU training, and multimodal models aligns with the role’s work.