Machine Learning Scientist

Are you an experienced Machine Learning Scientist with a desire to excel? If so, then Talent Software Services may have a job for you! Our client is seeking an experienced Machine Learning Scientist for a contract-to-hire role at their company in South San Francisco, CA. Position Summary: In the Structure and Simulation team within Prescient Design, we develop modern computational methods to accelerate therapeutic discovery across clients' Research and Early Development (gRED). Methods we deploy propose new molecules, score designs to prioritize the most promising compounds, generate biological hypotheses through exploratory simulation, accelerate physics-based calculations, and more. We are seeking a highly motivated Machine Learning Scientist to join our team to develop new scientific methodology and produce and deploy workflows usable by computational and, ultimately, wet-lab scientists. The successful candidate will collaborate extensively with computational and experimental researchers within Prescient Design and across gRED to advance our scientific understanding of biomolecules. Primary Responsibilities/Accountabilities: Work as a machine learning scientist to develop new scientific methodology for the understanding, scoring, ranking, generation, and design of biomolecules, especially proteins. Work as an engineer of scientific software, to produce usable, deployable code for these new methods to power the lab-in-the-loop. Use software best practices (version control, testing, modular code development, documentation, etc.) to collaborate on a large codebase with our team of methods developers. Deploy workflows on HPC and cloud platforms and deliver user-friendly web-based interfaces to medicinal chemists across gRED and Roche. Qualifications: BS, MS, or PhD degree in a life or physical science or a computational field. Expert in Python and experience with scientific software development. Experience with deploying software workflows on cloud and/or HPC platforms. Experience working on collaborative code bases, including merge requests, code review, writing tests etc. Highly motivated and independent self-starter who is eager to collaborate. Excellent communication and interpersonal skills. Basic understanding of modern machine learning methods, including predictive models, generative models, and active learning, as applied to molecular generation and optimization. Preferred: Candidates may additionally have, but are not required to have: Public portfolio of projects available on GitHub. Experience with Rosetta, OpenMM, and/or computational chemistry codes. 3+ years of industry experience. Extensive experience working with large chemical and biological datasets, including graph, sequence, and structure-based data.

Created: 2026-03-06