Data & AI Engineer (Semantic Layer)

Binariks is looking for a Data & AI Engineer (Semantic Layer) to design the knowledge that AI agents reason over. This covers how drugs, payers, coverage rules, prescribers and patient journeys connect, and the shared vocabulary every agent relies on.

The project is an emerging AI decision-intelligence platform for pharmaceutical commercial teams, developed for a US healthcare data and analytics company. It turns daily specialty-pharmacy, hub, payer and clinical data into AI-generated recommendations that show why they were made and who acts on them, then measures whether they worked. The role will join a senior Binariks team working alongside the client's product and engineering teams on AWS.

What we are looking for:

• 5+ years in data engineering, analytics engineering, or solutions engineering, with a solid track record in client-facing environments

• Experience integrating data/semantic context with LLM applications (RAG, GraphRAG, metadata-driven retrieval, or structured outputs) using Python

• Deep expertise in conceptual, logical, and physical data modelling - spanning ER, dimensional, and canonical models, with the ability to map complex business concepts directly to clean data structures

• Hands-on experience with modern data platforms: strong SQL, hands-on experience with AWS or GCP (AWS preferred), and working knowledge of Databricks (Delta Lake, Unity Catalog)

• Experience with data catalog and metadata management with tools such as DataHub, Collibra, Alation, Unity Catalog, Dataplex or Glue, including business glossaries and lineage

• Strong interpersonal skills to facilitate discovery workshops, translate domain jargon into robust technical designs, and articulate trade-offs clearly to non-technical stakeholders.

• Working proficiency in English (B2+)

• High comfort navigating ambiguity, rapid context-switching, and operating embedded directly within client organizations

Your Responsibilities:

• Design semantic models, business glossaries, and domain taxonomies to capture complex business logic, utilizing ontology standards (such as RDF/OWL or graph models) where applicable

• Implement semantic models (business entities, metric definitions) directly within Databricks Unity Catalog

• Design the foundational semantic contracts for AI agents (shared vocabulary, entity IDs, permission models, context scoping)

• Integrate knowledge graphs with LLM applications (implementing GraphRAG and semantic context retrieval)

• Build, automate, and refresh graph structures from underlying relational and lakehouse data sources

• Manage graph quality through validation, version control, lineage tracking, provenance, and data governance

• Facilitate discovery sessions with domain SMEs to capture business terms, logic, rules, and relationship mappings

Will Be a Plus:

• Practical background with OWL, RDF, SKOS, SHACL, or Protégé

• Experience building and querying graph databases (Neo4j, AWS Neptune, Stardog, GraphDB) via Cypher or SPARQL

• Hands-on work with GraphRAG

• Familiarity with regulated domain standards such as FHIR (Healthcare) or ACORD (Insurance)