Co-op, ML Scientist for Biology

Lila Sciences Lila Sciences · AI Frontier · San Francisco, CA · AI

Research role focused on developing autonomous-science capabilities in life sciences using AI and automation. The role involves exploring reasoning models, evaluating and reinforcing agentic model behavior, developing benchmark datasets, analyzing multi-modal biological data, and prototyping workflows connecting AI reasoning with scientific feedback.

What you'd actually do

  1. Contribute to ML research on reasoning models for biological discovery and autonomous science.
  2. Explore methods to evaluate, guide, and reinforce agentic model behavior in biological domains.
  3. Help develop evaluation and benchmark datasets for biological reasoning tasks.
  4. Analyze multi-modal biological data to identify useful signals for model evaluation and improvement.
  5. Prototype workflows that connect model reasoning, evaluation, and scientific feedback.

Skills

Required

  • Currently enrolled in a PhD program in Computer Science, Machine Learning, Computational Biology, Bioengineering, or a related quantitative field.
  • Research experience in machine learning, AI for science, computational biology, or biological data analysis.
  • Strong programming skills in Python and experience with modern ML frameworks such as PyTorch, JAX, or similar tools.
  • Experience working with biological, scientific, or multi-modal datasets.
  • Interest in reasoning models, agentic systems, evaluation methods, or benchmark design.
  • Interest in closed-loop scientific discovery, autonomous labs, or AI systems that interact with experimental feedback.
  • Ability to communicate research findings clearly through code, notebooks, written summaries, and presentations.
  • Comfort working in a collaborative, cross-disciplinary research environment.

Nice to have

  • Experience with reasoning models, agentic systems, reinforcement learning, or model evaluation.
  • Experience developing benchmarks, evaluation datasets, or model assessment workflows.
  • Publications, preprints, talks, posters, or workshop presentations in ML, AI for science, computational biology, or related scientific venues.

What the JD emphasized

  • autonomous-science capabilities
  • agentic model behavior
  • evaluation and benchmark datasets
  • multi-modal biological data
  • model reasoning

Other signals

  • autonomous-science capabilities
  • agentic model behavior
  • reasoning models
  • evaluation and benchmark datasets
  • multi-modal data
  • foundation models