Associate Scientist, Postdoctoral Fellow – Computational Protein Design

MSD•Cambridge, MA

1d•$82,000 - $92,000•Onsite

About The Position

Our company's Discovery Biologics is seeking a talented postdoctoral scientist motivated to push the boundaries of computational protein engineering and next-generation biotherapeutics design. In this role, you will develop and apply state-of-the-art artificial intelligence (AI) and machine learning (ML) tools in conjunction with physics-based modeling to address fundamental challenges facing computational protein design and structure prediction. The successful candidate will explore how ensemble-aware computational protein design (e.g. target conformational flexibility, state selectivity, multi-state binder design) can directly impact therapeutic pipelines and develop the methodology and tools to realize these unique advantages. Embedded in the Disruptive Technology group in our Cambridge, MA facility, the candidate will leverage our company's state-of-the-art capabilities and expertise in computational design, high-throughput production, and characterization for rapid data generation and model validation. As a postdoctoral fellow, your focus will be to deliver broadly relevant, high-impact science and computational models and tools. Publication of this work in peer-reviewed journals and scientific conferences is expected and the scope of the research project is designed specifically for external publication opportunity.

Requirements

Must currently hold a PhD OR Receive a Ph.D. no later than spring 2027
Ph.D. in Computational Biology, Bioinformatics, Computer Science, Biophysics, Computational Chemistry, Machine Learning, or a related quantitative discipline.
Hands-on experience developing and/or applying computational protein design tools (e.g., RFdiffusion, BoltzGen, ProteinMPNN) and structure prediction algorithms (e.g., AlphaFold, Boltz, ESMFold) for binder design.
Understanding of relevant experimental techniques for evaluation and characterization.
Quantitative evaluation mindset with a demonstrated ability to construct rigorous benchmarks, curate high-quality datasets, and apply critical-thinking and analytical skills for evaluation.
Strong proficiency in Python and modern ML frameworks (e.g., PyTorch, JAX), alongside experience with high-performance computing and reproducible software development.
Demonstrated commitment to collaboration in highly-interdisciplinary environments.
Strong track record of clear and effective scientific communication in addition to first-author peer-reviewed publications and presentations at major conferences.

Nice To Haves

Experience training or fine-tuning generative models (diffusion, flow-matching, etc.) or foundation models is highly desireable.
Experience with conformational ensemble generation and analysis — ensemble prediction methods (e.g., AF-cluster, AlphaFlow, BioEmu) and/or molecular dynamics tools (e.g., AMBER, Desmond, GROMACS, OpenMM).
Familiarity with energetic and free-energy methods for protein–protein interfaces (e.g., Rosetta/PyRosetta, Flex ddG, FEP).
Experimental experience is a plus but not required.
Familiarity with protein production, display technologies, and NGS-based selections with an understanding of how experimental data informs model development is advantageous.

Responsibilities

Implement rigorous benchmarking strategies for evaluating state-of-the-art computational protein design approaches.
Develop and validate ensemble-aware generative design and scoring methodologies for the advancement of next-generation biotherapeutics.
Communicate findings within our company and the broader scientific community by authoring high-impact manuscripts and distributing high-quality models and tools.