Job Description

At Lilly, we unite caring with discovery to make life better for people around the world. We are a global healthcare leader headquartered in Indianapolis, Indiana. Our employees around the world work to discover and bring life-changing medicines to those who need them, improve the understanding and management of disease, and give back to our communities through philanthropy and volunteerism. We give our best effort to our work, and we put people first. We’re looking for people who are determined to make life better for people around the world.

Position: Senior Agentic AI Engineer

Detailed Description:

At Eli Lilly and Company, we unite caring with discovery to make life better for people around the world. We are a global healthcare leader headquartered in Indianapolis, Indiana. Our employees work to discover and bring life-changing medicines to those who need them, improve the understanding and management of disease, and give back to our communities through philanthropy and volunteerism.

Eli Lilly is accelerating the adoption of agentic AI across its clinical development and health-economics technology landscape. We are seeking a Senior Agentic AI Engineer to build, deliver, and scale autonomous and semi-autonomous AI agent systems that transform how we manage clinical data, regulatory documents, health-economics evidence, and research operations. This individual brings deep, frontier-level knowledge of the latest developments across large language models, agentic frameworks, and AI tooling ecosystems — and is energized by experimentation, rapid prototyping, and translating cutting-edge research into production systems.

This is a hands-on engineering role. You will write production code, build agent orchestration pipelines, architect RAG and knowledge graph systems, and ship working software from Day 1. What makes this an R5-level position is not a shift away from building — it is the scope and impact of what you build. Your production implementations will become the reference architectures and platform standards that the broader organization adopts. Your code reviews and pairing sessions will be how you coach AI engineers and technical leads. The enterprise blueprints emerge from your working systems, not from slide decks.

The role spans a growing portfolio of high-impact work streams including clinical data automation, regulatory document intelligence, health-economics evidence platforms, and emerging AI-driven use cases across the clinical technology stack. The opportunity set is expanding rapidly, and this role is central to building the sustained engineering capacity Lilly needs to move at the pace the portfolio demands.

As a Senior Agentic AI Engineer, you will:

• Build and deliver multi-agent systems using agentic AI frameworks with support for tool-use, reflection, planning, and human-in-the-loop patterns across regulated pharmaceutical environments.
• Write production Python daily — designing LLM orchestration layers, RAG pipelines, knowledge graph integrations, and scalable MLOps infrastructure across AWS and Azure.
• Drive innovation in multi-agent systems and autonomous orchestration by developing novel approaches to complex technical problems, not by theorizing about them.
• Establish platform standards and reusable architecture patterns through reference implementations that others adopt — your shipped code becomes the blueprint.
• Coach AI engineers and technical leads through code reviews, pair programming, and hands-on problem-solving — elevating team capability by building alongside them.
• Accelerate delivery where external partners face capacity or pace constraints, operating as the go-to technical resource for agentic AI across the portfolio.
• Promote ideas and impact decisions across multiple teams and capabilities, challenging the status quo to improve engineering practices and drive innovation.
• Represent the team in enterprise AI architecture forums and communities of practice, grounding governance discussions in production realities.

Key Strategic Imperative: This role provides the critical hands-on engineering depth needed to scale Lilly’s agentic AI portfolio beyond what external delivery partners alone can sustain. The R5-level impact comes from building systems that set the standard — production implementations that become enterprise patterns, coaching that happens through shipping together, and innovation that is demonstrated in working software.

Key Responsibilities

Agentic System Design and Delivery (40%)

• Design, build, and deploy multi-agent systems using frameworks such as LangGraph, CrewAI, AutoGen, Semantic Kernel, or custom orchestration layers — owning the full lifecycle from prototype to production.
• Build agent workflows for clinical data lifecycle automation including study build, CDISC mapping (SDTM, ADaM), analysis dataset generation, TLF production, and regulatory submission packaging.
• Develop agentic pipelines for regulatory document management including intelligent classification, extraction, quality review, and lifecycle tracking of trial master file content.
• Build AI agent systems for health-economics and outcomes research (HEOR) use cases such as systematic literature review automation, evidence synthesis, network meta-analysis data extraction, and cost-effectiveness model parameterization.
• Drive innovation in autonomous orchestration and multi-agent patterns by building novel solutions — your working prototypes and production systems are how new approaches are validated and adopted.
• Deliver production-ready agent solutions across additional use cases as the portfolio expands, driving initiatives from concept to measurable business impact.

LLM Engineering, MLOps and Platform Integration (30%)

• Design and optimize LLM orchestration layers, prompt engineering strategies, multi-modal pipelines, and structured output parsing for domain-specific agent tasks.
• Design and build end-to-end RAG and agentic knowledge pipelines from scratch — including document ingestion (parsing, chunking, metadata extraction), embedding generation, vector database setup and management (Pinecone, Weaviate, Chroma, pgvector, FAISS), hybrid search configurations, and retrieval evaluation loops — tailored to clinical, regulatory, and health-economics corpora.
• Architect and build scalable MLOps and data pipelines across cloud platforms (AWS, Azure) including CI/CD for ML, model versioning, monitoring, and deployment automation.
• Build knowledge graph integrations and semantic search capabilities that enable agents to reason over complex, interconnected domain data.
• Develop APIs, microservices, and event-driven interfaces that expose agent capabilities to enterprise systems.
• Embed agentic AI capabilities into existing products, operations, and decision-making workflows across the clinical technology stack.
• Stay at the frontier of GenAI and agentic AI developments — actively tracking model releases, framework updates (LangGraph, AutoGen, smolagents, etc.), emerging protocols (MCP, A2A), and new agentic patterns — and rapidly experiment to evaluate applicability within the Lilly platform context.

Quality, Compliance and Validation (10%)

• Ensure AI agent outputs meet GxP, 21 CFR Part 11, and ICH regulatory standards; build validation-ready testing frameworks for agent-generated deliverables.
• Collaborate with Quality and CSV teams to implement agent validation strategies, audit trails, and explainability — baking compliance into the engineering process, not bolting it on after.
• Build guardrails, safety nets, confidence scoring, and human-in-the-loop checkpoints directly into agent systems.

Technical Leadership, Coaching and Influence (20%)

• Coach and mentor AI engineers and technical leads through code reviews, pair programming, architecture spikes, and hands-on problem-solving — not presentations.
• Design and lead a structured capability-building programme to grow a cohort of Agentic AI Architects within the organization — covering LLM fundamentals, model deployment patterns, agentic framework design, RAG pipeline construction, and evaluation methodologies — equipping participants with the technical depth to independently own and deliver agentic solutions.
• Establish platform standards and reusable components through reference implementations that the broader team and vendor partners adopt.
• Actively promote ideas and impact decisions across multiple teams and capabilities; challenge the status quo to improve engineering processes and drive innovation.
• Represent the team in enterprise AI architecture forums and communities of practice, ensuring governance discussions are grounded in production experience.
• Translate requirements from clinical programming, biostatistics, regulatory operations, and HEOR stakeholders into working agent systems, not just design specifications.

Strategic Execution Approach

This role is engineering-first. The strategic influence is a consequence of building well, not a separate activity:

Hands-On Engineering (60%)

Leadership Through Building (40%)

Build agent orchestration pipelines, RAG architectures, knowledge graph integrations, MLOps infrastructure. Write production Python. Design APIs. Implement guardrails. Deploy to cloud. Ship working software.

Establish platform standards through reference implementations. Coach engineers by pairing on production code. Drive architecture decisions grounded in what you have built and shipped. Represent the team in enterprise forums with production credibility.

Required Qualifications

Education

• Bachelor’s or Master’s degree in Computer Science, Artificial Intelligence, Data Science, Software Engineering, or a related quantitative discipline.

Experience

• 10+ years of software engineering experience with at least 4 years focused on AI/ML systems in production environments.
• 4+ years of hands-on experience building agentic AI systems (multi-agent orchestration, autonomous workflows, tool-use agents) — not limited to chatbot or single-prompt applications.
• Demonstrated track record of production implementations that became adopted platform patterns or reference architectures within an organization.
• Hands-on experience with at least two agentic frameworks (LangChain/LangGraph, CrewAI, AutoGen, Semantic Kernel, Haystack, or equivalent).
• Strong Python proficiency; working knowledge of TypeScript/Node.js for API layers.

Technical Depth (Hands-On)

• Deep expertise across the LLM stack: API integration and management (OpenAI, Anthropic, Mistral, Cohere, open-weight models via Ollama/vLLM/TGI), model deployment patterns for varied use cases (latency-sensitive inference, batch processing, edge/private deployment), prompt engineering, function calling, multi-modal orchestration, RAG, model distillation, structured output parsing, and rigorous evaluation and benchmarking of agent performance.
• Hands-on experience building scalable MLOps and data pipelines across cloud platforms (AWS, Azure) including CI/CD for ML, model versioning, monitoring, and deployment automation.
• Production experience building RAG architectures and knowledge pipelines from scratch: vector database setup and management (Pinecone, Weaviate, Chroma, pgvector, FAISS), document ingestion pipelines (parsing, chunking, metadata tagging, deduplication), embedding model selection and fine-tuning, hybrid search, and end-to-end retrieval evaluation and optimization.
• Experience building or integrating knowledge graphs for domain reasoning and semantic search.
• Proficiency with containerized deployments (Docker, Kubernetes), data flows across batch and streaming formats, and cloud-native AI services.
• Deep understanding of responsible AI principles: bias detection, hallucination mitigation, guardrail design, and explainability.
• Active, demonstrated knowledge of the latest developments in GenAI and agentic AI — including frontier model capabilities, emerging agentic frameworks, new protocol standards (MCP, A2A), agentic memory architectures, and multi-agent coordination patterns — with a proven appetite for rapid experimentation and translating research into working prototypes.

Leadership and Influence

• Track record of coaching and mentoring engineers through hands-on methods (code reviews, pairing, architecture spikes) that measurably increase team technical depth.
• Demonstrated ability to design and deliver structured technical training programmes that build a cohort of engineers with solid, applied knowledge of LLM model deployment, agentic system design, and AI platform architecture — not just conceptual awareness.
• Demonstrated ability to promote ideas and impact technical decisions across multiple teams — with influence earned through shipping, not seniority alone.
• Ability to communicate complex technical problems to non-technical audiences and ground strategy discussions in production realities.
• Proven ability to challenge the status quo by building better alternatives, driving improvements to engineering processes and practices.

Preferred Qualifications

• Experience in pharmaceutical, life sciences, or healthcare regulated environments (GxP, 21 CFR Part 11, HIPAA).
• Familiarity with CDISC standards (SDTM, ADaM), clinical trial data flows, regulatory submission processes, or trial master file management.
• Exposure to HEOR workflows: systematic literature reviews, meta-analyses, health-economic modelling (e.g., Markov models, microsimulation, budget impact analysis).
• Experience with SAS-to-R/Python migration in statistical computing contexts.
• Knowledge of MCP (Model Context Protocol), A2A (Agent-to-Agent) protocols, or emerging agent interoperability and communication standards.
• Prior work integrating AI agents with enterprise clinical or regulatory platforms (e.g., Veeva Vault, Medidata, SAS Drug Development).
• SAFe Agile or equivalent experience in scaled delivery environments; comfort operating within PI Planning and product-oriented delivery models.
• Experience shaping enterprise GenAI tooling, infrastructure, and guardrails through hands-on implementation.

Key Competencies

• Builder First: Writes production code daily. Moves from concept to working system rapidly. The keyboard is the primary tool, not the whiteboard.
• Specialization Through Depth: Extensive hands-on experience in agentic AI and autonomous orchestration; develops novel approaches by building and shipping them in production.
• Standards Through Delivery: Establishes platform patterns and architecture standards through reference implementations that others adopt — blueprints emerge from working systems.
• Coaching by Building Together: Elevates AI engineers and technical leads through code reviews, pair programming, and hands-on problem-solving, not slide decks.
• Multi-Team Influence: Actively promotes ideas and impacts decisions across multiple teams and capabilities; represents the organization in enterprise AI forums with production credibility.
• Challenges the Status Quo: Drives innovation by building better alternatives; does not accept ‘good enough’ when working software can prove a better way.
• Regulatory Awareness: Bakes compliance into the engineering process — builds auditable, traceable, validatable systems from day one.
• Ownership and Impact: Takes end-to-end accountability from prototype to production, driving initiatives to measurable business impact.

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.

#WeAreLilly