Computational Biology & ML Engineer

Computational Biology & ML Engineer – Peptide & Protein and chemical Design

Location: India (Remote/Hybrid)

Engagement: Full-time long-term contract

We are building an AI-enabled platform for designing and evaluating small molecules, peptide and folded protein binders. We are looking for a senior, hands-on individual contributor who can serve as both the technical lead and scientific subject matter expert for this initiative.

This is not a purely academic or software-engineering role. The successful candidate must be able to design the scientific approach, Use AI to output production-quality code, test the complete workflow, and evaluate whether the results are biologically meaningful.

Key responsibilities

• Build and validate end-to-end peptide and protein-design pipelines, from target preparation and candidate generation through structural verification and ranking.
• Integrate tools such as BoltzGen, Boltz-2, Protenix, Reinvent 4,P2Rank, inverse-folding models, and protein-interface analysis methods.
• Develop Python-based services for sequence generation, filtering, diversity selection, GPU-based inference, data capture, and reporting.
• Design, train, and benchmark target-conditioned ML models using distillation, uncertainty estimation, applicability checks, and active-learning methods.
• Establish scientific and software testing, including hard negatives, held-out target families, false-negative controls, reproducibility, throughput, and GPU-cost benchmarks.
• Act as the primary scientific advisor to product, engineering, and experimental teams.

Required background

• Advanced degree or equivalent experience in computational biology, bioinformatics, structural biology, machine learning, or a related field. Muster is a must.
• At least 2 years of organizational work experience is a must.
• Strong knowledge of small chemical molecules, peptide design, protein structure, protein-protein interfaces, and computational protein design.
• Python and practical experience with PyTorch or JAX.
• Experience working with Linux, Docker, GPUs, scientific data pipelines, APIs, and reproducible ML workflows.
• Ability to interpret structural confidence, PAE, binding-site coverage, contacts, clashes, developability, and candidate diversity.
• Demonstrated ability to independently build, test, and deliver scientific software.

Experience with BoltzGen/Boltz-2, AlphaFold-class models, Protenix, ProteinMPNN or other inverse-folding tools, PRODIGY, HPC/cloud GPU environments, and MLOps is highly desirable.