Director, Discovery Bioinformatics Oncology

Eli Lilly Eli Lilly · Pharma · San Francisco, CA

Lead the AI/ML innovation & deployment for oncology discovery, architecting and operationalizing state-of-the-art machine learning (deep learning, foundation models, LLM-powered applications) to accelerate target identification, protein/antibody design, and multimodal data integration. Develop next-gen data integration platforms, advance computational protein & antibody design using active learning, and design/oversee experiments. Deliver robust, scalable ML systems with MLOps on cloud platforms and foundational bioinformatics.

What you'd actually do

  1. Innovate and execute the AI/ML strategy for discovery. Build a portfolio of models for target ID/validation, structure‑ and sequence‑based protein design (e.g., antibodies, conjugates), mode‑of‑action inference, and biomarker discovery. Establish retrieval‑augmented and agentic LLM workflows for knowledge mining (literature, patents, internal reports) and protocol/screen design assistance.
  2. Develop next‑gen data integration platforms. Integrate bulk & single‑cell transcriptomics, WES/WGS, proteomics, CRISPR screen data, imaging, functional readouts, and real‑world knowledge graphs into unified model‑ready datasets. Drive ontology/harmonization, feature stores, and model registries for reproducibility and added value extraction.
  3. Advance computational protein & antibody design. Leverage transformer‑based sequence models, diffusion/graph methods, and physics‑informed constraints for binder optimization, specificity, and developability; operationalize active‑learning loops with design–make–test cycles. Lead antibody–siRNA conjugate design heuristics and predictive models for delivery and efficacy.
  4. Design & oversee experiments (dry & wet). Plan benchmarking and prospective validation; pair ML predictions with targeted assays and orthogonal analytics. Build feedback loops to refine models with experimental results and post‑market learnings.
  5. Cross‑functional impact & leadership. Partner with Biology/Chemistry/Translational/Clinical Biomarkers to convert insights into program decisions. Represent computational strategy in steering committees and external partnerships; publish/present at top venues. Mentor and grow a high‑performing team (data scientists, ML engineers, bioinformaticians) with strong engineering and scientific rigor.

Skills

Required

  • PhD or MS in Computer Science, Computational Biology, Bioinformatics, Statistics, Applied Math, or related STEM field.
  • 5+ years of post‑doctoral/industry experience delivering ML solutions in biotech/pharma or adjacent domains.

Nice to have

  • Experience in leading teams and cross‑functional initiatives is preferred.
  • Demonstrated impact applying deep learning to biological problems (e.g., transformers for protein/antibody sequence, structure prediction/refinement, graph learning, diffusion models, transfer learning, multimodal integration).
  • Deep hands‑on expertise with PyTorch (preferred) and/or JAX/TensorFlow; experience with Hugging Face (Transformers, Diffusers) and foundation‑model fine‑tuning (LoRA/PEFT, adapters, RAG).
  • Track record building LLM applications (prompt engineering, tool use/agents, vector databases, retrieval pipelines) for knowledge extraction, hypothesis generation, and protocol design in drug discovery.
  • Strong software engineering skills: Python, ML tooling (PyTorch Lightning, Hydra, Weights & Biases/MLflow), Git/GitHub, code review, testing & productionizing models with Docker/Kubernetes, APIs, and AWS services (e.g., S3, Batch/EKS, Lambda, Step Functions, SageMaker or equivalent).
  • Solid grounding in statistics/causal inference/experimental design; experience closing model–experiment loops.
  • Evidence of scientific leadership: high‑quality publications, patents, open‑source contributions, or conference talks.

What the JD emphasized

  • AI/ML innovation & deployment
  • state‑of‑the‑art machine learning
  • accelerate
  • protein and antibody design
  • multimodal data integration
  • transform heterogeneous molecular and phenotypic data into actionable hypotheses
  • design in silico–to–in vitro loops
  • deliver decision‑quality insights
  • platformization efforts
  • antibody, XDC development
  • next‑generation data products
  • Innovate and execute the AI/ML strategy
  • retrieval‑augmented and agentic LLM workflows
  • Develop next‑gen data integration platforms
  • Advance computational protein & antibody design
  • operationalize active‑learning loops
  • Design & oversee experiments
  • Build feedback loops
  • Cross‑functional impact & leadership
  • Mentor and grow a high‑performing team
  • Deliver robust, scalable ML systems
  • Foundational bioinformatics
  • Best‑practice omics analysis
  • applying deep learning to biological problems
  • foundation‑model fine‑tuning
  • Track record building LLM applications
  • Strong software engineering skills
  • Solid grounding in statistics/causal inference/experimental design
  • Evidence of scientific leadership

Other signals

  • AI/ML innovation & deployment
  • accelerate target identification & validation
  • protein and antibody design
  • multimodal data integration
  • transform heterogeneous molecular and phenotypic data into actionable hypotheses
  • design in silico–to–in vitro loops
  • deliver decision‑quality insights
  • platformization efforts for in silico design
  • antibody, XDC development
  • next‑generation data products