Research Scientist, Computational Biologist

Your impact

Design antibodies and other protein binders that turn difficult biological objectives into experimentally testable molecules.

As a Research Scientist, Computational Biologist, you will shape design strategies, develop scalable computational workflows, and interpret structural, sequence, and experimental evidence. Your work will influence which molecules we test, what we learn from them, and how our models and design methods improve.

You will work closely with AI researchers, experimental scientists, and software engineers. Because we are a small team, the role spans the full design loop: understanding the biological problem, generating and prioritizing candidates, analyzing experimental results, and turning what we learn into reusable capabilities.

What you will do

• Drive computational protein design campaigns. Translate biological and therapeutic objectives into design strategies. Prepare target structures, define epitopes and constraints, and generate candidates across individual targets and larger target panels.
• Develop and automate design workflows. Bring together generative models, sequence design, structure prediction, and molecular analysis in scalable, reproducible pipelines.
• Evaluate and prioritize designed molecules. Develop filtering and ranking strategies that account for structural quality, target accessibility, specificity, developability, and intended biological function.
• Build trustworthy scientific data foundations. Curate, validate, and connect sequence, structure, and experimental datasets for model training, evaluation, and discovery while preserving provenance.
• Learn from meaningful experimental outcomes. Benchmark methods, collaborate on screening and follow-up experiments, investigate gaps between predictions and measurements, and use the evidence to refine the next design round.
• Build capabilities the team can reuse. Turn recurring analyses into maintainable, tested tools and document the assumptions that other scientists and engineers need to use them well.

What you bring

You are a computational scientist who combines strong biological judgment with practical engineering. You can move from an evolving scientific question to a defensible analysis or design decision, and you know when to involve collaborators with complementary expertise.

• A Ph.D. in Computational Biology, Structural Biology, Biophysics, Bioengineering, Computer Science, or a related discipline, or equivalent research experience.
• Experience applying computational methods to protein design or structural modeling, with a strong understanding of protein structure–function relationships and protein–protein interactions.
• Hands-on experience with protein structure prediction or generative design models, such as AlphaFold, ESMFold, or RFdiffusion, and the judgment to interpret their outputs critically.
• Strong Python skills and experience building readable, tested, and reproducible scientific software.
• Experience processing biological data and defining quality criteria that are appropriate to the scientific question.
• Sound quantitative reasoning, clear communication, curiosity, and a habit of checking assumptions against evidence.

Depth in protein design, structural modeling, or biological data analysis matters more than experience with every tool. Strong scientific judgment and a willingness to build beyond your existing expertise are essential.

Experience that would help

• Antibody or de novo binder design, including CDR design, epitope targeting, interface analysis, specificity assessment, or developability evaluation.
• Curation of structural and sequence datasets for machine learning, including biological assembly assessment, annotation, clustering, and leakage-aware partitioning.
• Integration of computational design with high-throughput screening, binding measurements, cell-based functional assays, or affinity maturation.
• Cloud or high-performance computing, GPU workloads, containers, workflow orchestration, or effective use of coding agents for scientific work.

Our working environment

Our work combines Python-based scientific workflows, protein design and structural modeling tools, and sequence, structure, and experimental datasets. We use Docker and AWS alongside on-premises GPU infrastructure, working across disciplines to make computation scalable and reproducible.

Why Pythia

You will work directly with the people developing generative AI for molecular design and testing its output experimentally. Your scientific decisions will have a visible effect on the molecules we pursue and the evidence that guides our research.

We are a small, interdisciplinary team. We value people who take ownership, move between disciplines, check assumptions against evidence, and bring others into decisions. You will have room to shape both the work and how the team works.