Note: The job is a remote job and is open to candidates in USA. GHX is a healthcare business and data automation company that provides cloud-based supply chain technology, solutions, analytics, and services. The Senior Data Scientist - Ontology will design and maintain formal ontological architectures for cross-organizational data alignment, develop ontology mappings and validation processes, and direct LLM-assisted ontology discovery and enrichment. The role also collaborates with technical and business stakeholders to support transactional, clinical, and analytical requirements.
Responsibilities
- Design and maintain the ontology, covering the canonical structural layer (organizations, items, contracts, transaction), source data ontologies (supporting the canonical) and the process layer (data curation, ontology matching, workflows)
- Establish the rules for when two records from different systems refer to the same thing, and when they don't — recognizing the answer can differ by use case
- Establish mappings from trading partner source data to the canonical ontology, with documented provenance and validity conditions for each mapping
- Author OWL 2 axioms for ontology components; validate logical consistency (e.g. reasoner); maintain ontology lifecycle (e.g. with ROBOT, SHACL)
- Align with governance team and practice
- Grounded ontology discovery from data (and its uses) rather than schema declarations and metadata alone
- Build, direct and evaluate LLM-assisted ontology extraction pipelines, define and enforce the human-in-the-loop validation standards for AI-generated ontological candidates
- Collaborate with data quality engineers to establish formal feedback
- Translate formal ontology design decisions into specification/implementation for graph and relational stores
- Specify and implement SPARQL queries and graph schema requirements with sufficient precision to prevent implementation-level semantic loss
- Collaborate with internal and external stakeholders including domain experts, data engineers, product managers, and integration partners to ensure ontological architecture supports transactional, clinical, and analytical requirements
- Proactively monitor developments in formal ontology, knowledge representation, and LLM-assisted knowledge engineering to drive adoption of improved methods
Skills
- Greater than 4 years of experience in knowledge engineering, ontology development, or a closely related formal methods discipline
- Demonstrated experience building and maintaining domain ontologies in Protege or equivalent, with reasoner-validated consistency; not solely taxonomy or metadata management work
- Experience with ROBOT or ODK for ontology lifecycle management (or similar): automated quality checks, versioning, release pipelines
- Expertise in SPARQL and/or Cypher for querying ontology-aligned data stores; ability to write and evaluate queries that correctly reflect ontological intent
- Demonstrated ability to interpret data profiling output and translate it into formal ontological claims; experience with empirical ontology discovery from data as well as top-down ontology design
- Experience directing or evaluating LLM-assisted knowledge extraction pipelines with formal validation requirements
- Proficiency in Python (or similar) for ontology tooling, pipeline scripting, and data analysis in support of knowledge engineering workflows
- Experience working in multi-disciplinary teams where formal and domain knowledge must be integrated under operational constraints
- Bachelor's or advanced degree in Computer Science, Mathematics, Philosophy (logic/formal methods), Information Science, or a related hard science discipline
- Familiarity with category theory as applied to data integration -- functors, natural transformations, limits and colimits as schema merge operations -- at literacy level or above; knowledge of CQL/AQL or categorical database theory is a plus
- Experience with LinkML
- Healthcare supply chain domain knowledge and ontological structures
- Experience with BFO 2.0 and the OBO Foundry principles and standards
- Familiarity with provenance models (why-provenance, how-provenance, where-provenance) and their implementation in ontology-aligned data systems
- Experience with graph database platforms at production scale (Stardog, Amazon Neptune, or equivalent) and the operational considerations of ontology-driven graph deployments
- Passion for staying at the cutting edge of knowledge representation, semantic alignment, and AI-assisted ontology engineering
- Sense of humor
Company Overview
Company H1B Sponsorship