Computational Protein Design Expert

Job Description – Pod Lead, Computational Protein Design Expert (Terminal Bench Science)

Budget: $14 per hour

Role Overview

We are seeking experienced Computational Protein Design experts to explore and evaluate advanced generative AI approaches for protein engineering as part of a frontier AI scientific evaluation initiative.

The role involves developing challenging scientific evaluation tasks related to de novo protein generation, diffusion-based structure design, protein binder development, and sequence–structure optimization. You will conduct computational experiments, evaluate AI-generated protein designs, and develop reproducible benchmarks and technical proof-of-concept results to assess the capabilities and limitations of generative AI models in protein engineering.

Key Responsibilities

• Design scientifically rigorous evaluation tasks involving de novo protein design, target-conditioned binder generation, motif scaffolding, and sequence–structure optimization.
• Develop and execute computational protein design workflows using generative AI models, diffusion-based approaches, and established protein engineering tools.
• Define task inputs, biological constraints, expected outputs, evaluation criteria, and reproducible experimental protocols.
• Evaluate AI-generated protein sequences and structures for folding confidence, structural stability predictions, geometric validity, interface quality, novelty, and biological plausibility.
• Analyze protein folding, sequence–structure relationships, and protein–protein interactions to assess the quality and feasibility of generated designs.
• Develop objective scoring rubrics, reference analyses, and computational validation methods to evaluate model performance.
• Identify model limitations, failure cases, unrealistic structural predictions, and potential weaknesses in generated protein designs.
• Distinguish computational predictions from experimentally validated results and document scientific assumptions and limitations.
• Prepare reproducible evaluation benchmarks, computational analysis reports, and technical documentation.
• Summarize findings in concise technical reports and contribute to proof-of-concept demonstrations for scientific AI evaluation initiatives.

Required Skills & Qualifications

• Ph.D. (ongoing or completed) in Computational Biology, Structural Biology, Bioinformatics, Biochemistry, Protein Engineering, or a closely related discipline.
• Hands-on experience in at least one major area: computational protein design, protein structure prediction, molecular modeling, or protein engineering.
• Strong understanding of protein folding, protein structure–function relationships, molecular interactions, and sequence–structure optimization.
• Practical familiarity with computational protein design and structural analysis tools such as RFdiffusion, ProteinMPNN, AlphaFold, Rosetta, PyMOL, or ChimeraX.
• Strong programming skills in Python, with experience developing or executing computational biology workflows.
• Experience working with protein sequence and structural data formats such as FASTA, PDB, and mmCIF.
• Ability to evaluate protein structures, interpret computational predictions, and assess the scientific validity of generated designs.
• Strong analytical and problem-solving skills, with the ability to independently design experiments, evaluate results, and identify technical limitations.
• Excellent written communication skills for documenting scientific findings, evaluation methodologies, and technical reports.

Nice to Have

• Experience with generative AI or diffusion-based protein design models, particularly RFdiffusion or similar approaches.
• Hands-on experience in de novo protein design, protein binder design, motif scaffolding, or enzyme engineering.
• Experience with protein sequence optimization, structural refinement, protein–protein docking, or computational affinity assessment.
• Familiarity with computational screening, protein design benchmarking, and automated evaluation pipelines.
• Experience developing scientific evaluation tasks, scoring rubrics, reference solutions, or reproducible computational benchmarks.
• Experience evaluating AI-generated biological structures, scientific foundation models, or generative protein design systems.
• Research experience involving computational protein engineering, molecular simulations, or structural bioinformatics.
• Publications, research contributions, or open-source development experience in computational biology or protein design.
• Prior experience in AI training, scientific model evaluation, technical proof-of-concept development, or research collaborations.

Expected Deliverables

• A curated collection of reproducible computational protein design evaluation tasks, including clearly defined inputs, expected outputs, and objective scoring rubrics.
• Computational analysis of AI-generated protein designs, documenting structural quality, biological plausibility, model limitations, and failure cases.
• A concise technical report summarizing evaluation findings, supported by a proof-of-concept demonstration for a potential scientific AI evaluation partnership.