Data Scientist -AI Drug Discovery

Overview

Data Scientist in AI Drug Discovery at Lilly Bengaluru, using data science and machine learning to answer drug discovery questions, with a core focus on single-cell biology and virtual cell modelling.

What you'll do

• Build, train, benchmark and validate machine learning models for scientific data, including single-cell omics, genetics, chemistry and imaging.
• Select methods based on the scientific question, from classical machine learning to deep learning and foundation models.
• Develop and evaluate models that predict cellular state and response to genetic or chemical perturbations, and use them for in silico screening and prioritising high-value experiments.
• Combine model outputs, perturbation evidence and prior biological knowledge to nominate and rank candidate targets.
• Communicate strength of evidence, uncertainty, limitations and proposed experimental validation.
• Build reproducible workflows to ingest, harmonise, quality-control and document public and internal datasets across formats and modalities.
• Incorporate regulatory and interaction networks, ontologies, pathways and other biological or chemical priors into models, and quantify the contribution of each information source.
• Run training and analysis on HPC, GPU and cloud infrastructure using versioned code, containers, automated testing and reusable software components.
• Partner with scientists and engineers to frame questions, define decision criteria, and interpret findings, and present clearly to technical, scientific, and leadership audiences.

What you'll need

• Master's degree in Machine Learning, Computer Science, Computational Chemistry, Computational Biology, Physics, Statistics, or a closely related quantitative field.
• 3+ years of experience applying machine learning in industry or academia, including delivering end-to-end projects and contributing to research or technical direction within a team.
• Proficiency in Python and modern ML frameworks such as PyTorch, with hands-on experience developing and evaluating models.
• Strong Python skills and practical experience with machine learning and deep learning networks.
• Solid grounding in machine learning, model validation, and experimental design.
• Working knowledge of transformers or graph neural networks and of sound train/validation/test design.
• Experience working with large, noisy or heterogeneous scientific and biomedical datasets.
• Experience with single-cell omics, genetics, chemistry or imaging data.
• Experience running training and analysis on HPC, GPU or cloud infrastructure.
• Ability to communicate technical findings, uncertainty, and limitations to collaborators from different disciplines.

Nice to have

• Experience with single-cell and perturbation data (e.g., Perturb-seq, CRISPR screens) and virtual cell or foundation models.
• Familiarity with statistical genetics methods (GWAS, fine-mapping, colocalization) and multi-omics integration.
• Experience with knowledge graphs, ontologies, pathway databases, or other structured biological priors.

Details

• Location: Bengaluru, India.