Principal Machine Learning Scientist

Bayer•Cambridge, MA

About The Position

The Principal Machine Learning Scientist will develop novel machine learning algorithms and workflows for accelerating early-stage drug discovery. In this role, you are responsible for constructing, studying, and training algorithms that learn from complex, high-dimensional data to uncover patterns and develop practical predictive models and applications. Involves utilizing various techniques, such as random forests, deep learning, and neural networks, to enhance the predictive capabilities of algorithms, particularly in natural language processing and machine perception. Focuses on simulating human learning activities, improving system performance through data analysis, and developing deep learning frameworks and systems that operate independently of explicit programming instructions. By continuously refining models and exploring new methodologies, contributes to innovative solutions that leverage machine learning for diverse applications.

Requirements

Ph.D. degree in Computational Chemistry/Biology, Chem/Bioinformatics, Chemical/Biological/Molecular Engineering, or a related field at the intersection of life sciences and computer science
Deep expertise with state-of-the-art machine learning methods for modeling biomolecules, like co-folding and/or generative methods for protein design
Expertise in handling, processing, integrating and analyzing large datasets related to drug development research, including biochemical, biophysical, and structural biology data
Strong programming skills in Python
Demonstrated commitment to scientific rigor, a track record of scientific excellence, strong analytical thinking, and a high degree of self-motivation
Excellent written and verbal communication

Nice To Haves

5+ years of relevant post-PhD experience, including 2+ years in industry
Experience with established, physics-based protein modelling methods like Molecular Dynamics and/or Rosetta
Experience in coordinating small, interdisciplinary teams and ability to articulate their impact to managerial stakeholders
Strong record of publications or patents related to machine learning solutions for biomolecular modeling

Responsibilities

Develop, evaluate, and apply machine learning algorithms and workflows for accelerating early-stage drug discovery, including but not limited to (i) de-novo design of biomolecules, (ii) assessment of target druggability across therapeutic modalities (iii) design of drug delivery systems, (iv) identification of novel druggable pockets and epitopes, (vi) characterization of protein-protein and protein-ligand interactions
Contribute to the implementation, validation, and improvement of machine learning tools and software solutions that support drug discovery activities
Identify opportunities for accelerating ongoing drug discovery projects with internal and external AI capabilities
Communicate, educate, and engage with a broad set of stakeholders (chemists, biologists, computational/data scientists, R&D leadership) on the state of technology and the progress of key internal initiatives. Engage with the broader scientific community through publications, talks, and open-source
Keep up to date with the latest advances in AI-driven modeling of biomolecular structure and dynamics.