Research Engineer/Scientist, Imitation Learning & Pretraining
Tau Robotics· Research
Hybrid (San Francisco)full-timePosted 2mo ago
About the role
Research engineer/scientist role focused on imitation learning and pretraining for robotics, involving large-scale multimodal data, multi-task manipulation policies, and finetuning within an RL loop.
Responsibilities
- Build and scale supervised pretraining runs on large multi-modal robot datasets
- Train multi-task manipulation policies and study what actually transfers
- Define the finetuning recipes that hand pretrained policies to the RL loop
Qualifications
- Experience with large-scale pretraining, behaviour cloning, or vision-language-action models
- Comfort building data pipelines for large multimodal datasets
- Strong Python plus PyTorch or JAX
- Evidence: papers, repos, or models you trained that others used
Skills
Required:Large-scale pretrainingData pipelines for multimodal datasetsPythonPyTorch or JAX
Last updated Oct 17, 2026. Interested? Apply on the company site.
About Tau Robotics
Tau Robotics: Humanoid Cleaning Service in San Francisco
San Francisco’s Favorite Humanoid Cleaning Service
House CleaningDeep CleaningOffice CleaningVacuumingTaking out the trashClearing the clutter
More open roles at Tau Robotics
- Research Engineer/Scientist, World Models2mo agoHybrid (San Francisco) · Research
- Research Engineer, Simulation & Evaluation2mo agoHybrid (San Francisco) · Research
- Robotics Hardware Engineer, Prototyping & Integration2mo agoSan Francisco · Hardware
- Robotics Technician, Assembly & Maintenance2mo agoSan Francisco
- Robotics Software Engineer, Systems & Connectivity.2mo agoHybrid (San Francisco, CA) · Software