Materials Knowledge Architect

About the role

Join cuspai as a Materials Knowledge Architect to lead the design of scientific data representations and ingestion pipelines. This role is crucial for transforming diverse external data into high-quality, actionable assets for materials discovery. You will work with leading experts to build the data infrastructure that powers our AI models.

Key facts

What you'll do

• Design data models for chemical, structural, and materials property information, ensuring interoperability across various external sources.
• Establish and maintain frameworks for data quality, validation, deduplication, and lineage to ensure all ingested data is reliable and ready for machine learning.
• Apply expertise in chemistry, physics, or materials science to improve the quality, scale, and consistency of incoming data.
• Lead data ingestion from partners, building pipelines to normalize and harmonize data from diverse formats and standards.
• Help identify and evaluate new data sources and partnerships, including computational and experimental campaigns.
• Collaborate directly with external data providers to understand their data, define schema mappings, and resolve quality issues.
• Work with research and engineering teams to translate modeling needs into ingestion requirements and ML-ready datasets.
• Communicate your work and suggest improvements for data models, ingestion processes, and overall data quality.

Requirements

• Demonstrated experience designing data models and schemas for complex scientific or technical fields, and curating high-quality data assets.
• Proven experience ingesting, integrating, and harmonizing data from multiple external sources with varying formats and quality.
• PhD in Chemistry, Physics, Materials Science, Computational Chemistry, or a related scientific research field, or equivalent experience.
• Expert knowledge of data representation, ontologies, data modeling, and curating high-quality, interoperable data assets.
• Experience with a wide range of data types used in materials discovery, including experimental (e.g., synthesis, XRD, spectroscopy) and computed properties (e.g., DFT energies).
• Deep knowledge of materials and chemical data sources (e.g., ICSD, Cambridge Structural Database, NOMAD, Materials Project) and challenges in harmonizing lab-generated data.
• Working knowledge of Python and SQL, with experience building data pipelines for experimental and computational data, including schema design (e.g., Pydantic, JSON Schema) and familiarity with materials and data science toolkits (e.g., pymatgen, ASE, Pandas/Polars).
• Firsthand understanding of how experimental data is produced, including lab workflows and instrument quirks, and practical strategies for accurate modeling.
• Strong communication and collaboration skills, comfortable working with external partners and multidisciplinary teams.

Nice to have

• Industry experience at a materials, chemicals, energy, or deep-tech company, or close collaboration with industry in a research setting.
• Familiarity with data engineering concepts like ETL/ELT pipelines, schema validation, data contracts, workflow orchestration (e.g., Airflow, Dagster), and cloud infrastructure.
• Experience defining data-sharing agreements, data standards, or interchange formats with external partners.
• Experience with high-throughput computational screening workflows and DFT codes (e.g., VASP, Quantum ESPRESSO).

Skills & tools

• Python
• SQL
• Pydantic
• JSON Schema
• pymatgen
• ASE
• Pandas/Polars
• Airflow (nice to have)
• Dagster (nice to have)
• VASP (nice to have)
• Quantum ESPRESSO (nice to have)

Practical notes

This role can be based in our Cambridge, London, Amsterdam, or Berlin offices, with an expectation of three days per week in the office. Regular travel to other locations may be required. Benefits include 28 days holiday (UK, DE, NL), 26 weeks primary caregiver parental leave, 12 weeks secondary caregiver parental leave, and a professional development budget. cuspai is an equal opportunities employer and encourages applications from all backgrounds.