5 months ago
London, United KingdomMid Level
Responsibilities
- Post-train manipulation policies using behavior cloning and reinforcement learning, owning the workflow from data to deployment.
- Develop data-preprocessing strategies and improve the quality, diversity, and coverage of collected robotics data.
- Set up reinforcement-learning training with the simulation team and improve rewards and simulation quality for real-world transfer.
- Define data requirements and instructions for a specific robot capability in partnership with the data-collection organization.
- Expand robot observation and action spaces and collaborate with teleoperations teams to expose new components to operators.
- Improve motion smoothness and the teleoperation experience with Teleoperations and Controls teams.
- Communicate manipulation findings to the hardware design team for future robot designs.
Requirements
- At least 3 years of industry or research experience working on robots, with shipped artifacts.
- Good understanding of modern teleoperation and low-level control stacks.
- Experience with neural-network post-training.
- Familiarity with streaming datasets, checkpointing and state management, distributed training, PyTorch or JAX, numerical profiling and debugging, and maintainable research code.
- Familiarity with modern software engineering practices and the ability to document experiments and communicate trade-offs clearly.
- Preferred: experience training VLA models for manipulation using autoregressive, diffusion, or flow-matching approaches.
- Preferred: familiarity with OpenVLA, Physical Intelligence (π) models, or similar open VLA frameworks.
- Preferred: experience applying reinforcement learning to robotics.
- Preferred: publications at top-tier robotics or deep-learning conferences or equivalent open-source contributions.
Benefits
- Competitive equity through stock options with meaningful upside.
- 30+ paid days off, including 23 days of annual leave, UK bank holidays, and additional company closure days.
- Private healthcare with virtual and in-person care.
- Pension scheme with an 8% total contribution, including 5% employee and 3% employer contributions, on full earnings.
- Free daily breakfast, catered lunch, and in-office snacks.
- London-based work with daily collaboration in the office.
- Opportunity to work with engineers, researchers, and product experts on humanoid robotics and AI.
- Significant ownership, access to founding leadership, and input into product direction from day one.
Tech Stack
Categories
ML EngineeringRobotics
