About PolyModels Hub

PolyModels Hub is building the Modeling Operating System for the pharmaceutical industry. ModelFlow empowers scientists and engineers to unify computational modelling, machine learning, AI and scientific workflows, accelerating the development of life-changing therapies across biologics, small molecules and advanced modalities.

As ModelFlow evolves into an enterprise workflow platform, the data and knowledge layer becomes central: connecting enterprise data sources, workflow runs, models, artifacts, outputs and scientific concepts into a structured system that can support traceability, search, reuse and AI-assisted workflow creation.

Role Summary

We are hiring a Senior AI & Data Platform Engineer to help build the data foundation for ModelFlow. This role will support reliable connectivity to customer data environments while helping define the metadata, lineage and ontology structures that make scientific workflows reusable, searchable, auditable and AI-ready.

You will work across data engineering, semantic modelling and product integration to ensure that scientific data, models, workflow runs, artifacts and outputs can be connected, structured, queried, governed and reused across the platform.

What You Will Own

1. Data Connectivity & Integration

• Design and implement connectors to enterprise data systems, including Databricks, Snowflake, databases and file-based sources where needed.
• Build reliable patterns for data ingestion, query execution, schema discovery, data preview, export/import and output persistence.
• Ensure integrations are secure, scalable and suitable for enterprise customers.
• Translate customer-specific data access needs into reusable platform patterns rather than one-off integrations.

2. Data Modelling & Metadata

• Define practical data structures for datasets, workflow runs, artifacts, models, parameters, scenarios, outputs and lineage.
• Standardise how data and outputs are represented across workflows so they can be reused, searched and governed.
• Ensure generated outputs are structured and queryable where appropriate, rather than treated only as files or blobs.
• Create metadata patterns that support model versioning, execution context, operator attribution and reproducibility.

3. Ontology & Knowledge Layer Support

• Contribute to the ontology strategy for ModelFlow, working with domain experts, product and engineering.
• Help define relationships between scientific problems, workflows, models, datasets, parameters, artifacts and results.