Senior ML Scientist, Biological Systems

Your Impact at LILA

Lila is redefining the future of medicine by combining automated large-scale data generation with scientific superintelligence. At Lila, we are building the loop where AI, automation, and experimental biology co-evolve to solve the hardest problems in medicine.

The Life Science AI team develops machine learning systems for automated reasoning on biological data, combining state-of-the-art ML with breakthrough biology. We seek a Senior ML Scientist to focus on domain models underneath that system: models of perturbation biology, genetics, and and high-dimensional experimental readouts that connect biological mechanism, experimental intervention and therapeutic opportunity.

At Lila, the data is generated for the model: we run our own experiments at scale, and you will help decide what gets measured. The question is not only what to learn from a fixed dataset, but what datasets should exist. And model outputs are reasoned over. Predictions go to systems and scientists choosing what to run next and making mechanistic calls downstream. Both need to know why, not only what. A model that returns a quantity a biologist, human or AI, can argue with; models that transfer to new contexts, compose into a downstream calculation, and can be checked against an independent measurement are often worth more than a more accurate one that returns an embedding.

What You'll Be Building

Build domain models for perturbation, genetic, and multimodal experimental data, and translate biological questions into rigorous ML problem formulations. Where the biology supports it, build models with interpretable structure — representations grounded in the machinery that executes a biological process rather than in cell-type identity embeddings, and parameters a biologist can argue with.

Make uncertainty a deliverable. Calibrated posteriors, honest error bars, and outputs that tell a downstream consumer how much a prediction can actually settle — particularly when predicting into conditions never measured.

Pre-register and beat baselines. Simple entity-mean, additive, and linear baselines are stated before any model is fit. If the full model does not beat them, we ship the baseline and say so.

Partner with experimental scientists to guide data generation and model validation — shaping what gets measured, in what contexts, at what precision — and build the experiment-selection methods that choose the next batch of measurements to reduce uncertainty where it matters.

Design benchmarks that connect model performance to biological and therapeutic consequence, apply model outputs to prioritize targets and mechanisms, and contribute technical direction to high-impact modeling programs.

Support the integration of domain models into agentic workflows. Deploy tools, help engineer the agent harnesses that incorporate these tools and collaborate on creating environments to train the reasoning models that drive these agents.

What You'll Need to Succeed

PhD in machine learning, statistics, computational biology, computer science, bioengineering, physics, or a related quantitative field, with a strong publication record or equivalent industry impact.

Experience developing models for high-dimensional biological data, and the judgment to connect ML methods to biological mechanism and experimental design.

Generalization under structured sparsity. A track record of building models that predict into conditions not directly observed — sparse or unbalanced experimental designs, held-out combinations, transfer to new contexts — rather than interpolating within a densely sampled corpus.

Uncertainty and evaluation rigor. Comfort with calibration, proper scoring rules, and evaluation design, in work where someone made a decision based on your numbers.

Strong programming skills, reliable ML research workflows, and clear communication across ML, biology, and experimental teams. Track record of leading ambiguous research problems from …