AI Scientist

About Proxima (formerly VantAI):

Proxima is building a computational pipeline combining state-of-the-art physics-based modeling and machine learning to revolutionize drug discovery and development. Working together with some of the world’s leading biopharmaceutical companies, we design, test, and optimize novel therapies to treat some of the world’s most difficult diseases.

About You:
We are looking for an experienced AI Scientist to join our machine learning team to help develop the world’s most advanced pipeline for the design of proximity-inducing molecules. You will work with a team of world-class machine learning engineers in an interdisciplinary, research-heavy position on a range of unsolved problems around representation learning of proteins, small molecules, biological networks and genomics.

Key Responsibilities:

• Scientifically direct the design and training of large-scale, state-of-the art Deep Learning systems

• Develop novel architecture and training paradigms to lead the industry in unsolved scientific problems

• Collaborate with content experts from other domains (e.g., chemistry, physics, biology) to enable innovative feature-engineering and novel cross-disciplinary approaches

• Actively contribute to top-tier ML conferences and journals and attend core ML conferences to stay connected with the community and current trends

Basic Requirements:

• MS/PhD degree in Computer Science, Statistics, Applied Mathematics, Computational Biology, Computational Chemistry or other related subject (will also consider BS degrees in these areas for candidates highly qualified across other requirements or with significant work experience)

• Track record of contributing to novel methods for state-of-the-art Deep Learning (in industry or through publications)

• 2+ years of experience on machine learning teams, ideally at a start-up

• 4+ years of ML research experience in industry or academia, with strong familiarity with PyTorch

• Experience with Python is required

• Relevant experience working in a Linux/UNIX environment with basic data engineering and scripting abilities

• Ability to understand business problems and how to build models that can quickly drive value, while prioritizing your research efforts accordingly

• Clearly communicate modeling setup and results at various levels of abstraction

Preferred Qualifications:

• Domain knowledge and experience in Cheminformatics, Bioinformatics, Genomics, Computational Biology, Biophysics or Computational Chemistry

• Competitive programming or scientific experience, including Kaggle, PUTNAM, CTFs, iGEM, Biology/Chemistry Olympiad

• Strong working knowledge of containerized production (Docker, Kubernetes), DevOps, MLOps and CI/CD principles

• Programming skills in Rust, C, C++ is a plus

• Experience with other state-of-the-art frameworks such as Jax, TensorFlow, MXNet or Sklearn

• Experience working with large datasets in cloud environment with distributed computing tools (e.g., Spark, Airflow, Dask)

• Machine learning or software engineering project management experience