Senior Machine Learning Researcher - MSR AI for Science at Microsoft

Cambridge, England, United Kingdom -

Full Time

Start Date

Immediate

Expiry Date

06 Mar, 26

Salary

0.0

Posted On

06 Dec, 25

Experience

5 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

Machine Learning, Deep Learning, Python, PyTorch, Statistics, Linear Algebra, Model Design, Algorithm Design, Data Pipelines, Biomolecular Modeling, Bioinformatics, Chemoinformatics, Computational Chemistry, Geometric DL, Reinforcement Learning, Generative Models

Industry

Software Development

Description

Overview At Microsoft Research AI for Science , we believe machine learning and artificial intelligence has the potential to transform scientific modelling and discovery crucial for solving the most pressing problems facing society including sustainable materials and discovery of new drugs. We seek a highly motivated ML Researchers to join our Biomolecular Emulator (BioEmu) and Small Molecules Drug Discovery project. The BioEmu project aims to model the dynamics and function of proteins --- how they change shape, bind to each other, and bind small molecules. Our BioEmu-1 model was published in Science (see also our blog post)). In our Small molecules - Microsoft Research work we build and apply large ML and LLM algorithms to accelerate the discovery of small molecule drugs and materials. Our team encompasses people from multiple disciplines across machine learning, engineering, and the natural sciences, who work together closely on well-defined and challenging goals. If you have strong machine learning expertise and enjoy designing and creating tools for scalable machine learning research for the natural sciences, please apply. Responsibilities Invent novel deep learning techniques for models of (bio)molecular structure, dynamics, reactivity and function. Design, implement, and iterate on model architectures and training algorithms (e.g., diffusion/sequence–structure models, language models, representation learning); run rigorous ablations and baselines. Define success where standards don’t exist yet: proposing sound benchmarks and uncertainty‑aware metrics that reflect real‑world utility. Build high‑quality research code (Python/PyTorch) with reproducible workflows and robust data pipelines. Partner across disciplines—communicate clearly with ML researchers and experimental/computational biologists/chemists; present results and influence direction. Work autonomously and as a team player, reporting insights, risks, and next steps with crisp written/visual summaries. Thrive with imperfect, heterogeneous data, using principled curation, augmentation, and probabilistic evaluation. Aim for impact: try ideas quickly and fail-fast when they don't work. Rapidly convert working ideas to artifacts others can use (code, models, datasets, papers, patents). Qualifications Required/Minimum Qualifications: PhD or equivalent research experience in Computer Science, Machine Learning, Physics, Chemistry, or a related field. Demonstrated leadership in ML architecture and algorithm design. Strong expertise in deep learning (model design, large-scale training, evaluation and reproducibility), statistics and linear algebra. Proficiency in Python and modern ML/scientific frameworks (e.g., PyTorch, JAX, TensorFlow, NumPy, SciPy, Pandas). Peer-reviewed publications in leading venues (e.g., NeurIPS, ICML, ICLR or leading journals). Excellent technical communication for collaborating in an interdisciplinary team. Curiosity and drive to apply deep learning to problems in biology or chemistry. Comfort with real‑world, noisy/heterogeneous data. Preferred/Additional Qualifications: ML Engineering skills (e.g., model optimization and deployment, code design, CUDA). Experience with geometric DL, reinforcement learning, generative models or large language models. Experience with biomolecular modeling or bioinformatics (e.g., folding systems, structural analysis/visualization, MD simulation, structure/genome databases) and/or chemical modelling (2D/3D structures), chemoinformatics and computational chemistry. Ability to work with and interpret real‑world biological data (e.g., cryo‑EM, protein binding affinities, structural/biophysical/chemical measurements, molecular data). #Research #AI for Science This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled. Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, citizenship, color, family or medical care leave, gender identity or expression, genetic information, immigration status, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran or military status, race, ethnicity, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable local laws, regulations and ordinances. If you need assistance with religious accommodations and/or a reasonable accommodation due to a disability during the application process, read more about requesting accommodations.

Responsibilities

The role involves inventing novel deep learning techniques for modeling biomolecular structures and dynamics, as well as designing and implementing model architectures. The researcher will also collaborate across disciplines and communicate findings effectively.