Scientist II, Applied Machine Learning

Role Description

A-Alpha Bio is seeking a Scientist II to join our Applied Machine Learning team in turning large-scale protein interaction data into insights and models that advance antibody and protein engineering. In this role, you'll lead computational work for both internal research and external collaborations, from defining scientific questions and analysis plans to developing models and communicating results. Working with AlphaSeq datasets, you'll connect experimental measurements with protein sequence and structural information to understand binding, evaluate computational design methods, and guide future experiments.

Key Responsibilities

• Lead scientific analyses: Define and execute analysis plans for large-scale protein interaction datasets (often 1M PPIs), including evaluating binding affinity, specificity, and the performance of antibody and protein design methods.
• Develop and evaluate ML models: Build, train, and apply models using experimental measurements, protein sequences, and structural information. Design rigorous benchmarks and assess model performance, uncertainty, and generalization.
• Connect computation with experiments: Collaborate with experimental scientists to shape study designs, interpret results, investigate unexpected findings, and recommend follow-up experiments.
• Deliver results to external partners: Serve as the computational lead for assigned projects, producing clear presentations, reports, and well-documented datasets for both internal teams and external collaborators.
• Build reusable methods: Translate project-specific work into reliable analytical tools and workflows. Contribute maintainable Python code, review contributions, and document methods so others can use them.
• Advance our scientific capabilities: Identify and develop new computational approaches that increase the value of AlphaSeq data for protein engineering.

Qualifications & Experience

Required

• PhD in a relevant field with 2+ years of experience working on large-scale data analysis using biological sequence or structure data; or a bachelor's degree with 10+ years of relevant experience, including 5+ years working on large-scale data analysis using biological sequence or structure data; or an equivalent education and experience.
• Demonstrated expertise in Python for scientific computing, data analysis, visualization, and ML model development.
• Experience analyzing large experimental datasets and making scientifically defensible decisions about data quality, statistical methods, and interpretation.
• Strong understanding of model evaluation, including appropriate baselines, data leakage, and the limits of conclusions drawn from a benchmark.
• Ability to translate an open-ended scientific question into an executable plan and deliver reproducible results with limited supervision.
• Clear communication skills, including the ability to explain methods, findings, and limitations to computational and experimental audiences.
• Experience contributing to a shared codebase through version control, code review, and documentation.

Preferred

• Experience with antibody or protein engineering, particularly the relationship between sequence, structure, and binding.
• Familiarity with the structural modeling of protein complexes and evaluation of structure prediction or generative design methods.
• Use of modern ML frameworks such as PyTorch, TensorFlow, or Hugging Face.
• Experience processing and analyzing next-generation sequencing data (Illumina, Oxford Nanopore, etc.).
• Experience developing analytical tools or workflows used by other scientists.