Lead AI Engineer, Knowledge & Semantic Systems

Bausch+Lomb Companies Inc.•Bridgewater, NJ

1d•$150,000 - $200,000•Hybrid

About The Position

The Lead AI Engineer, Knowledge & Semantic Systems is a highly technical individual contributor responsible for building and maintaining the foundational data, knowledge, and retrieval systems that power enterprise AI solutions. The role creates trusted semantic models, knowledge graphs, metadata frameworks, and retrieval capabilities to support accurate, scalable, and compliant AI-driven insights. It partners with data engineering, AI, analytics, and business teams to strengthen enterprise AI capabilities and improve decision-making and business performance. The position owns the governed commercial semantic layer and commercial ontology, including canonical metric definitions, entity and relationship modeling, taxonomies, identity resolution, and temporal validity. It also owns the retrieval stack, evaluates retrieval quality, manages production vector infrastructure, and exposes semantic, graph, and retrieval capabilities to agents through well-specified interfaces. The engineer collaborates closely with Applied AI Agents engineers and other stakeholders, diagnoses grounding failures, curates contributions to the canonical model, and enforces governance, privacy, compliance, and auditability at the definition and retrieval layers. The role follows a hybrid schedule of up to two days per week telecommuting from home and three days per week in the Bridgewater, New Jersey corporate office. Applicants must be authorized to work for any U.S. employer; visa sponsorship is not available. The stated starting pay range is $150,000 to $200,000. U.S.-based employees may be eligible for incentives and the listed benefit programs.

Requirements

Bachelor's or Master's degree in Computer Science, Data Engineering, Information Science, Machine Learning, or a related field.
At least 5 years of experience in data engineering, analytics, semantic modeling, or knowledge engineering.
Expertise in SQL, dimensional data modeling, and business metric governance.
Hands-on experience with semantic layer tools such as dbt Semantic Layer, MetricFlow, Cube, LookML, or AtScale.
Experience designing and managing knowledge graphs, ontologies, and graph databases such as Neo4j.
Experience with entity resolution, identity management, and master data concepts.
Experience with vector databases, retrieval systems, and search optimization.
Understanding of LLM-enabled solutions, including RAG, tool calling, structured outputs, and MCP.
Strong Python development skills and software engineering best practices, including CI/CD, testing, version control, and containerization.
Experience with cloud data platforms and tools such as Snowflake, Databricks, AWS or Azure, dbt, Airflow, or Dagster.
Knowledge of data governance, security, lineage, and compliance requirements.
Ability to collaborate effectively across business and technical teams.
Experience working in Agile environments using Scrum, SAFe, or similar methodologies.
Excellent communication and documentation skills, including the ability to align stakeholders on data definitions and standards.
Authorization to work for any U.S. employer; the employer is unable to sponsor or take over sponsorship of an employment visa.

Nice To Haves

Experience with pharmaceutical commercial data, including prescription, claims, payer, specialty pharmacy, hub, and CRM data.
Familiarity with eye care industry dynamics, including ophthalmology, optometry, and Eye Care Professional networks.
Experience with customer master data management (MDM) and HCP/HCO data in life sciences.
Experience with graph databases such as Neo4j and vector database technologies.
Familiarity with content metadata, classification, and approval workflows.
Experience with data governance, cataloging, lineage, and data observability tools.
Knowledge of healthcare compliance requirements, including HIPAA and FDA promotional regulations.
Experience working in regulated industries with strong audit and compliance standards.

Responsibilities

Design, build, and maintain the governed commercial semantic layer, including canonical metric definitions, dimensions, hierarchies, joins, data grain, and row- and role-level access policies.
Codify commercial metrics such as TRx, NRx, NBRx, NC TRx, market share, market basket, decile and segment assignments, and projected versus captured volume, documenting source systems, calculations, refresh cadence, and known limitations.
Resolve conflicting metric definitions across brand, field, market access, and finance stakeholders.
Enforce constraints that prevent semantically misleading queries, including incompatible volume types, grains, or non-comparable sources.
Design and extend the commercial ontology and knowledge graph across ECPs, accounts, affiliations, payer and plan data, formulary status, products, territories, content assets, and engagement.
Apply entity resolution across source systems and reconcile prescriber and account identities across prescription, claims, specialty pharmacy, and CRM feeds; maintain group practice and affiliation structures.
Maintain taxonomies, controlled vocabularies, synonyms, and aliases so varied business language resolves to governed concepts consistently.
Encode temporal validity for territory realignments, formulary changes, affiliation changes, and segment refreshes to support historically accurate answers.
Own the retrieval stack end to end, including chunking strategy, embedding model selection and evaluation, and index configuration and tuning.
Build labeled query sets and measure retrieval quality using recall and precision at k, groundedness, and citation accuracy; implement regression testing for corpus, embedding, or chunking changes.
Manage reindexing, embedding drift, model migration, and vector infrastructure cost and performance at production scale.
Expose semantic, graph, and retrieval capabilities to agents through defined interfaces, including MCP servers and resources, tool contracts, and structured schemas.
Partner with Applied AI Agents engineers on a shared sprint cadence and support agent answer accuracy and delivery velocity.
Diagnose incorrect or unsupported agent answers, identify whether the cause is definitions, entity resolution, retrieval, or model reasoning, and remediate issues within this layer.
Review proposed metrics, entities, and relationships from analysts and agent engineers, and curate a consistent canonical model.
Partner with Data Engineering, IT, Privacy, and Compliance to enforce access policies, PDRP and HIPAA minimum-necessary handling, and audit trails at the definition and retrieval layers.
Document and socialize the model and provide enablement so analysts and engineers adopt governed definitions instead of rebuilding them locally.