SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Lila Sciences is seeking a Senior Machine Learning Scientist to develop data-efficient learning strategies for drug discovery. This applied research role focuses on building ML models and experimental design approaches for settings where data is scarce, expensive, and intentionally generated—typically datasets ranging from tens to low thousands of high-quality examples in focused chemical spaces.
You will design and implement approaches across active learning, meta-learning, fine-tuning, uncertainty estimation, experimental design, and multimodal modeling to enable closed-loop learning systems. The work bridges model training with scientific decision-making: your data acquisition recommendations will inform computational chemists evaluating compound priorities, computational biophysicists deciding when simulation is warranted, and structure-aware modelers selecting which protein-ligand data would improve predictions.
Key responsibilities include: building ML models that perform well in low-data regimes for molecular optimization; designing data acquisition strategies to identify which compounds, assays, simulations, or experiments should run next to maximize learning; developing active learning and uncertainty-aware approaches for focused chemical spaces; training on low-quantity, high-quality datasets from Lila's experimental and computational systems; building multimodal models integrating DEL data, simulation outputs, assay data, structural information, chemical features, and experimental metadata; partnering with experimental and drug discovery teams to ensure recommendations are scientifically meaningful and operationally feasible; evaluating models through learning curves, prospective validation, and decision-focused metrics; developing closed-loop workflows that continuously update as new data arrives; and translating model predictions into practical recommendations for compound and assay selection.
Required: PhD or equivalent in machine learning, computational chemistry, computational biology, statistics, computer science, or related field. Strong experience training ML models in low-data regimes and familiarity with active learning, Bayesian optimization, experimental design, meta-learning, transfer learning, or uncertainty estimation. Experience building ML models for scientific, molecular, or biological datasets. Multimodal learning experience. Ability to reason about data acquisition strategy, not just model fitting. Strong scientific judgment connecting model behavior to experimental decisions. Practical experience with PyTorch, JAX, or scikit-learn. Ability to collaborate across ML, data, computational science, and drug discovery teams.
Bonus: drug discovery experience, especially in molecular optimization or design-make-test-learn workflows; understanding of pharmacology or biochemistry; experience with DEL, high-throughput screening, medicinal chemistry, or simulation-derived features; closed-loop experimentation or autonomous lab experience; generative molecular design or batch selection workflows; familiarity with causal inference, optimal experimental design, or Bayesian methods.
