Cheminformatics & Synthesis Prediction

Develop and improve reaction-enumeration workflows, including reaction SMARTS templates, functional-group gating, building-block curation, and production runs.

Work with large vendor catalogs while balancing chemical coverage, price, availability, lead time, and data quality.

Improve treatment of regioisomers, stereochemistry, resolution limits, and other sources of ambiguity in enumerated chemical space.

Build reliable workflows for structure handling, reaction execution, sanitization, identifiers, SDF files, and metadata.

Integrate and evaluate approaches for retrosynthesis, synthetic success, reaction-condition prediction, and related reaction modeling tasks.

Assess models and workflows for calibration, coverage, applicability domain, and practical usefulness; surface uncertainty and risk flags rather than bare point estimates.

Help connect predicted reactions and enumerated compounds to experimental design, make–test workflows, and downstream learning.

Consolidate reaction data—including conditions, yields, failed reactions, and provenance—into a shared, machine-readable source that supports future model improvement.

Write maintainable code and contribute to service-oriented systems, deployment workflows, and data pipelines.

Partner with internal chemists and cross-functional teams to translate scientific questions into reliable computational workflows and interpret results.

Scope and review external or consultant work with clear specifications and acceptance criteria.

Work on commercial digital products by integrating pricing and ordering data and functionalities, and deploying and maintaining related services and tools.

Design, build, and optimize Model Context Protocols (MCPs) and agentic frameworks to support LLM-based work and integrate cheminformatics tools into automated workflows.

Ph.D. in cheminformatics, computational chemistry, organic chemistry, medicinal chemistry, or a closely related quantitative field, plus 3 years of relevant industry or postdoctoral experience; or an M.S. with 6 years, or a B.S. with 9 years, of relevant experience.

Strong practical experience in cheminformatics and reaction prediction, including experience with retrosynthesis, reaction-condition prediction, synthetic accessibility, or related workflows.

Experience working with reaction SMARTS, functional-group classification, molecular representations, and chemical structure data.

Production experience with RDKit or an equivalent cheminformatics toolkit.

Experience working with large chemical or vendor building-block catalogs and understanding the trade-offs involved in coverage, scale, and data quality.

Strong Python skills and experience writing code that others can run and maintain.

Familiarity with service-oriented software, Docker, environment-based configuration, and data pipelines.

Working knowledge of common medicinal-chemistry synthetic transformations and the ability to collaborate effectively with synthetic chemists.

Clear technical communication skills and a disciplined approach to documentation, provenance, and reproducibility.

You understand that models are mathematical representations of the current state of knowledge: they will never be perfect, but they can support learning and better decisions. You know how to distinguish useful signal from an output that is not reliable enough to use, and communicate uncertainty and limitations clearly.

You balance scientific rigor with practical progress. You make thoughtful trade-offs, avoid overclaiming, and collaborate clearly with chemists and engineers to turn imperfect computational tools into useful workflows.

You are intellectually curious and impatient with the status quo, with a passion for establishing and advancing state-of-the-art computational methods in practical production settings.

Experience with large makeable or synthesizable chemical spaces, whether commercial or in-house.

Experience operating a combinatorial library-enumeration pipeline or reaction-prediction workflow in production.

Experience evaluating predictive models, including uncertainty estimation, calibration, applicability domain, or ranked candidate generation.

Experience with regiochemistry, stereochemistry, chemical registration systems, ELN/LIMS integration, or chemical data standards.

Experience connecting computational design to wet-lab results in a design–make–test–learn environment.

Contributions to open-source cheminformatics or experience evaluating third-party chemistry platforms.

Experience building and optimizing Model Context Protocols (MCPs), agentic frameworks, or integrating cheminformatics tools with LLM-based workflows.

Build cheminformatics technology and workflows on Nebula, the world’s largest autonomous laboratory.

Help connect molecular design, reaction prediction, synthesis, testing, and learning across Ginkgo’s automated platforms.

Develop workflows to ingest, structure, and learn from the high-quality data generated by autonomous laboratory systems.

Join a technically ambitious, cross-functional team building the computational foundations for AI-enabled drug discovery.

Receive competitive compensation, generous equity in a public company, and a robust benefits package.

The base salary range for this role is $115,00-165,000. Actual pay within this range will depend on a candidate's skills, expertise, and experience. We also offer company stock awards, a comprehensive benefits package including medical, dental & vision coverage, health spending accounts, voluntary benefits, leave of absence policies, 401(k) program with employer contribution, 8 paid holidays in addition to a full-week winter shutdown and unlimited Paid Time Off policy.

Ginkgo has implemented a return to office policy effective October 1, 2025 with required in office days 5x per week. This policy applies to all employees who live within 50 miles of Ginkgo’s offices in Boston, MA, Emeryville, CA and West Sacramento, CA.