C
Materials Knowledge Architect
Contract type
Ongoing
Work mode
Hybrid · Cambridge, United Kingdom · London, United Kingdom
Experience
3 years
Job description
A Materials Knowledge Architect at CuspAI's Data team who designs scientific data representations, schemas and ingestion pipelines for materials and chemical data. The role owns how heterogeneous external data is modelled, harmonised and made ML-ready.
At a glance
- Employment
- Full-time · Employee (employment agreement)
- Work mode
- Hybrid · Cambridge, United Kingdom · London, United Kingdom
What you'll do
- Design the data models that represent chemical, structural and materials property information, and define how diverse external sources map onto a coherent, interoperable internal representation.
- Establish and own data quality, validation, deduplication, lineage and provenance frameworks so that every ingested dataset is trustworthy, traceable and ML-ready.
- Use expertise in chemistry, physics and/or materials science to maximise the quality, scale and consistency of data flowing in from external sources.
- Lead the ingestion of data from partners — industrial collaborators, instrument and simulation vendors, research institutions and commercial data providers — building robust pipelines that normalise, harmonise and reconcile data arriving in varied formats and standards.
- Contribute to the data team's efforts to identify, evaluate and assess new data sources, partnerships and data generation opportunities, including computational (e.g. DFT, molecular dynamics) and experimental campaigns.
- Partner directly with external data providers to understand their data, agree formats and standards, define schema mappings, and resolve quality, provenance and interoperability issues at the source.
- Work in partnership across research and engineering teams to translate modelling needs into ingestion requirements and ML-ready datasets.
- Communicate work and raise awareness of opportunities to improve data models, ingestion processes and overall data quality.
What we're looking for
- Proven experience designing data models and schemas for complex scientific or technical domains, and curating high-quality data assets from them.
- Demonstrable experience ingesting, integrating and harmonising data from multiple external sources or partners, each with differing formats, schemas and quality levels.
- PhD in Chemistry, Physics, Materials Science, Computational Chemistry or a related discipline, or equivalent experience in scientific research.
- Expert in data representation, ontologies, data modelling and the curation of high-quality, interoperable data assets.
- Experience working with a broad range of data types used in materials discovery — especially experimental and characterization data (e.g. synthesis parameters and processing conditions, XRD patterns, spectroscopy, adsorption isotherms, device measurements) alongside computed properties (e.g. DFT-derived formation energies, band gaps) — and mapping them across heterogeneous sources.
- Deep knowledge of materials and chemical data sources spanning both experimental and computational domains (e.g. ICSD, the Cambridge Structural Database, NOMAD, Materials Project), and of the practical challenges of harmonizing lab-generated data — instrument outputs, ELN/LIMS records, unit conventions, provenance — into interoperable, ML-ready assets.
- Working knowledge of Python and SQL with experience building data pipelines for experimental and computational data, including schema/data-model design (e.g. Pydantic, JSON Schema) and familiarity with materials and data science toolkits (e.g. pymatgen, ASE, Pandas/Polars).
- Firsthand experience with how experimental data is actually produced — lab workflows, instrument quirks, incomplete metadata — and pragmatic strategies for modeling it faithfully rather than forcing it into idealized schemas.
- Strong communicator and a proven collaborator, comfortable working directly with external partners as well as multidisciplinary chemistry/physics/materials and product/engineering teams.
Nice to have
- Experience working in industry at a materials, chemicals, energy or deep-tech company, or closely with industry at a research institution or national lab.
- Familiarity with data engineering concepts — ETL/ELT pipelines, schema validation, data contracts, workflow orchestration (e.g. Airflow, Dagster) — and experience running jobs on cloud-based 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).
Your team
Data team
The data team at CuspAI, working closely with world-leading AI experts and materials science researchers.
Benefits
- A competitive salary
- Equity in CuspAI
- 28 days holiday (DE, NL, UK) or 21 days holiday (JP, SG, US), in addition to local public holidays
- 'Gold Standard' parental leave: 26 weeks (primary caregiver) and 12 weeks (secondary caregiver) at full pay
- Professional development budget
- Solve meaningful problems
- True interdisciplinary teamwork
Apply
Direct apply at Cuspai
Opens the job posting at Cuspai
About CuspAI
The frontier AI company on a mission to solve the breakthrough materials needed to power human progress
CuspAI is the frontier AI company building a 'search engine' for next-generation materials, working on challenges including energy, clean water, the future of compute, and carbon capture. Its founding team is comprised of world-class researchers in AI, chemistry and engineering.
Mission
To solve the breakthrough materials needed to power human progress by harnessing AI to unlock trillion-dollar materials breakthroughs in months, not millennia.
Key facts
- AI / materials science
- Founding team is the most cited in the world, comprised of world-class researchers in AI, chemistry and engineering
- Working on energy, clean water, the future of compute, and carbon capture
- Building a 'search engine' for next-generation materials
What they value
- Diverse and inclusive workplace
- Collaboration
- Interdisciplinary teamwork