About the job
Recursion Pharmaceuticals is reimagining drug discovery by combining biology and data in new ways. The Senior Scientific Data Engineer - Data Platform will help shape and maintain the data infrastructure that drives this mission. This position is based in either London or Oxford.
Role overview
This role focuses on designing, developing, and maintaining scientific data systems that underpin Recursion’s research and product development. The emphasis is on data architecture rather than machine learning or model building. The work involves:
- Managing the ingestion, standardization, and distribution of public and proprietary datasets essential for drug discovery.
- Enabling competitor intelligence, chemical tractability analysis, and compound design workflows through well-structured data products.
- Owning and evolving the data infrastructure that supports predictive modeling and machine learning across the company.
This position suits engineers who enjoy building complex scientific data systems. It is not a modeling or data science role.
Key data systems and products
- Flagship SAR Data Mart: Integrates commercial and public bioactivity databases (such as ChEMBL) with internal assay results.
- Commercial Vendor Data Mart: Maintains a catalog of purchasable compounds for internal design tools and tractability assessments.
- Biomedical Knowledge Graph: Offers semantic graph infrastructure connecting targets, diseases, and compounds to support AI-driven research.
- Chemical Synthesis Data: Stores reaction datasets for training retrosynthesis models and predicting chemical tractability.
- Patent Intelligence System: Converts patent feeds and competitor data into actionable insights for research and strategy.
- Compound Standardization Registry: Maintains a large-scale chemical structure repository to ensure consistency across billions of compounds, similar to UniChem.
Important note
This is a specialized Data Engineering position focused on data infrastructure and stewardship. The work does not include training or building predictive models, but it directly supports those efforts through reliable systems and curated datasets.
