PR&D – Technical Lead Architect – Semantic Layer & Agentic AI

At Lilly, the work is demanding because patients are waiting. We unite caring with discovery to help make life better for people around the world, knowing that every decision, every detail, and every day matters. Headquartered in Indianapolis, Indiana, our over 50,000 employees around the globe take on complex challenges to discover and deliver life-changing medicines, strengthen how health is understood and managed, and support the communities we serve. This is hard, urgent, selfless work—but it’s work worth doing. If you’re driven by purpose and ready to bring your best to work that truly matters for patients, we invite you to join us. 

Product Research & Development (PR&D) · CTI-MD Tech@Lilly

Product Research & Development (PR&D) owns the CMC (Chemistry, Manufacturing, and Controls) data that underpins every regulatory submission, batch investigation, and process decision across Lilly's development portfolio, spanning structured systems (LIMS, MES, DMS, ELN) and unstructured scientific content (protocols, batch records, investigation reports). The function operates under strict GxP, data integrity, and audit-readiness requirements. The Technical Lead Architect (R5), PR&D, builds and owns the Semantic Layer and Knowledge Graph that convert fragmented, siloed CMC data into a unified, ontology-driven foundation usable by both humans and AI agents — and the agentic layer that lets PR&D's agents reliably retrieve, reason over, and act on that foundation. This role is hands-on and implementation-heavy — building ontologies, triplets, knowledge graphs, and agent orchestration directly — while also carrying business-facing accountability for the Semantic Layer roadmap and technical leadership of the engineers who build alongside them. This role ensures the Semantic Layer remains a PR&D-owned strategic asset, maximizing reuse across data products, reducing fragmentation, and creating measurable business value through faster, more reliable AI-assisted scientific workflows.

Path/Level: R5 (Technical Lead Architect)

Key Responsibilities

• Unified Data Foundation — map structured, semi-structured, and unstructured CMC/PR&D assets into a connected Knowledge Graph and Vector Database, using LLM-assisted tooling to accelerate ontology development and reduce manual curation.
• Multi-Ontology Design & Ownership — design and maintain multiple connected ontologies — a structured ontology and a business ontology — as PR&D-owned artifacts.
• Triplet & Knowledge Graph Engineering — own end-to-end knowledge graph construction: entity/relationship extraction, triplet creation, triplet aggregation, triplet validation, and graph population across Neo4j/GraphDB and vector stores (Azure AI Search, AWS OpenSearch, or equivalent).
• Multi-Modal Data Ingestion — build ingestion pipelines handling multi-modal and multi-lingual PR&D content (documents, structured records, and beyond text-only formats where relevant), with ontology-based tagging and entity/intent recognition feeding both the knowledge graph and vector store.
• Retrieval Optimization — optimize hybrid graph, vector, and keyword retrieval for both human exploration and agentic tool calls, tuning for precision, recall, latency, and groundedness across KBQ, metadata, and document-level questions answered from a single semantic layer.
• Ontology-Driven Tagging & Metadata Enrichment — apply domain-specific ontologies for semantic tagging and contextual indexing of PR&D documents; build technical and business metadata enrichment pipelines (table/column descriptions, join conditions, business rules) using LLMs.
• GenAI & Agentic Architecture — design and deploy an agentic architecture (Q&A agent, summarization agent, context retriever, and similar) exposed via MCP (Model Context Protocol) servers, enabling fast, relevant, and explainable knowledge retrieval for scientific workflows.
• Agent Orchestration & Operational Guardrails — design orchestration patterns for PR&D's knowledge and domain agents to interoperate (agent-to-agent handoffs, not just agent-to-tool via MCP); establish evaluation frameworks, observability/monitoring, and a prompt library for agents operating against the Semantic Layer, ensuring PR&D's agentic footprint stays governable as it scales beyond single-agent use cases.
• Business Partnership & Roadmap Ownership — partner directly with PR&D business and science stakeholders to translate scientific workflow needs (e.g., batch investigations, CMC submissions) into Semantic Layer priorities; own and communicate the technical roadmap to senior stakeholders, framing trade-offs in business rather than purely technical terms.
• Technical Leadership & Delivery Oversight — provide hands-on technical leadership to a team of data/knowledge graph engineers — setting technical direction, unblocking design decisions, and reviewing implementation quality, while remaining a primary builder rather than a pure people manager.
• Governance & Compliance — establish architecture review processes, ontology change management, and validation/traceability frameworks ensuring the Semantic Layer meets FDA 21 CFR Part 11, GxP, ICH, and enterprise security and data privacy requirements (HIPAA, GDPR, SOC 2, ISO 27001).
• Technology Evaluation & Rationalization — evaluate emerging knowledge graph, vector database, and agentic AI technologies for CMC applicability; run staged (Bronze/Silver/Gold) proof-of-concepts before enterprise-wide adoption.
• Reusable Architecture Patterns — establish reusable ontology, triplet, and retrieval design patterns and reference implementations that accelerate delivery for domain teams building on the Semantic Layer.
• Talent Development & Mentorship — mentor data engineers, platform engineers, and architects on ontology design, knowledge graph engineering, and retrieval optimization practices, fostering technical depth across PR&D.

Technical & Strategic Expertise Areas

Knowledge Graph Engineering & Optimization · Ontology Design & Knowledge Modelling · Triplet Creation, Aggregation & Validation · RDF/OWL/SPARQL & Property Graphs (Neo4j, AWS Neptune) · Vector Databases & Embeddings (Azure AI Search, AWS OpenSearch, Pinecone) · Hybrid & Graph-RAG Retrieval · Unstructured-to-Structured Data Conversion · Multi-Modal & Multi-Lingual Data Ingestion · Generative AI & Agentic AI · Multi-Agent Architecture · Agent-to-Agent (A2A) Protocols · Agent Design Patterns (ReAct, Plan-and-Execute, Reflection) · MCP (Model Context Protocol) · Agent Evaluation Frameworks · Agent Observability & Monitoring · Prompt Library/Hub Management · Advanced Prompt & Context Engineering · Data Architecture & Data Product Development · Cloud Platforms (Azure/AWS) · Containerisation & Kubernetes · Security & Compliance Architecture · API & Integration Architecture · Python · Stakeholder Management & Roadmap Communication · Technical Team Leadership · Vendor & Delivery Governance

Basic Qualifications

• Proven track record building production knowledge graphs and semantic layers end to end — not overseeing them.
• Demonstrated success implementing complex semantic layers spanning both structured and unstructured data.
• Experience driving ontology/knowledge graph technology decisions, retrieval optimization, and architecture governance.
• Experience operating multiple agents in production with shared guardrails, evaluation, and observability tooling — not just single-agent RAG/Q&A systems.
• Experience partnering with business/science stakeholders to shape technical roadmaps and prioritization.
• Track record providing technical leadership to engineering teams while remaining hands-on.
• Track record of successfully leading through ambiguity, complexity, and organizational transformation.

Additional Qualifications / Competencies

• Experience in CMC, pharmaceutical, or life sciences data (batch records, process parameters, LIMS/MES/DMS/ELN systems).
• Familiarity with data-to-ontology mapping tooling and document-to-schema conversion frameworks.
• Experience exposing knowledge graphs to LLM/agent consumption via MCP or equivalent tool-calling interfaces.
• Experience designing agent-to-agent (A2A) communication patterns and multi-agent orchestration beyond single-agent tool calling.
• Comfortable presenting technical strategy and trade-offs to senior, non-technical stakeholders.
• Background in pharma and GxP-regulated environments.

Required Leadership Capabilities

• Able to go deep technically while translating graph/retrieval trade-offs into business-relevant narratives for senior stakeholders.
• Comfortable owning and defending a technical roadmap in front of business and executive audiences.
• Provides day-to-day technical leadership and mentorship to engineers without requiring a formal people-manager structure.
• Systems thinking connecting ontology/graph design choices to downstream agentic workflow and business outcomes.

Lilly is dedicated to helping individuals with disabilities to actively engage in the workforce, ensuring equal opportunities when vying for positions. If you require accommodation to submit a resume for a position at Lilly, please complete the accommodation request form (https://careers.lilly.com/us/en/workplace-accommodation) for further assistance. Please note this is for individuals to request an accommodation as part of the application process and any other correspondence will not receive a response.

Lilly does not discriminate on the basis of age, race, color, religion, gender, sexual orientation, gender identity, gender expression, national origin, protected veteran status, disability or any other legally protected status.