Machine Learning Research Intern (Summer 2026)

Tahoe Therapeutics • South San Francisco, CA

22h • Onsite

About The Position

Tahoe Therapeutics is a biotechnology company pioneering a fundamentally new approach to drug discovery, one that begins with the biology of real patients. Our Mosaic platform is the first to make in vivo data generation scalable, with single-cell resolution, allowing us to map how drugs affect patient-derived cells in the body across a wide range of biological contexts. We are building the world’s largest in vivo single-cell perturbation atlas and using it to train multimodal foundation models that learn the context-dependent nature of gene function, disease progression, and drug response.
By combining cutting-edge machine learning with the most biologically relevant datasets ever assembled in drug discovery, our mission is to find better drugs, faster and bring them to more patients who need them.
Your role
With Tahoe-100M, we solved one of the fundamental bottlenecks in building a virtual model of the cell: generating massive, perturbation-rich, single-cell datasets that capture real biological causality. With Tahoe-x1, we removed the second bottleneck: creating a modern platform for rapid iteration on model architectures and designs in a cost-efficient manner and at scale. At Tahoe, we embody a simple philosophy: build in the open, shoot for the moon, and we’re looking for people who want to push the frontier of virtual cell models.
As a Machine Learning Research Intern, you will join our ML team for ~16 weeks over the summer to develop and evaluate perturbation prediction models on our large-scale single-cell datasets such as Tahoe-100M and beyond. You will work on-site full-time at our South San Francisco office.

Requirements

• Currently enrolled in a degree (undergraduate, Master's, or PhD) in CS, ML, computational biology, or a related field.
• Strong fundamentals in deep learning with hands-on experience in PyTorch, JAX, or TensorFlow. Systems level familiarity with model throughput optimization and GPU kernels is a plus.
• Experience training modern deep learning architectures (Transformers, diffusion models, state-space models, etc.). through research or course projects.

Nice To Haves

• Systems level familiarity with model throughput optimization and GPU kernels is a plus.
• Exposure to ML for biological data is a plus, but not required.

Responsibilities

• Work closely with the team to advance the state-of-the-art in deep learning models for perturbation effect prediction.
• Contribute across our full ML stack: data processing, model training, evaluation, leaderboards, and external APIs.

Benefits

• Work directly with a world-class ML team on high-impact research, with access to cutting-edge compute
• Daily lunch

What This Job Offers

• Job Type: Full-time
• Career Level: Intern