Impress employers and recruiters.
Choose from hundreds of resume examples.
Impress employers and recruiters.
Choose from hundreds of resume examples.
Target Resume Now
About the Role
Anvil is seeking its first robot learning hire to implement cutting-edge research on their hardware, shipping polished demos and guides. This role offers significant leverage over data collection and hardware specifications, with direct access to manufacturing and factory resources.
Responsibilities
- Own training pipelines from data ingestion to evaluation harness.
- Replicate high-leverage public research into published demos and guides.
- Validate and improve the UMI data collection to policy deployment pipeline.
- Develop data flywheel workflows for policy improvement and scaled collection.
- Ensure all shipped projects meet a high standard of polish, including video, guide, and reproducible repo.
Requirements
- Personally trained and deployed imitation-learning policies on real robot arms.
- Fluent in the layer beneath the model, including action chunking, temporal ensembling, inference latency, and synchronization.
- Ability to replicate papers and transfer research to hardware within weeks.
- Strong data instincts to identify issues in demonstrations.
- Proven ability to finish projects with polished deliverables (guide, video, repo).
- Honest reporting of results, including metrics and failure modes.
- Comfortable operating as the sole ML person in a fast-paced startup environment.
- Experience with ACT, Diffusion Policy, or VLA fine-tunes.
- Understanding of action chunking, temporal ensembling, inference latency, and control-loop frequency.
- Proficiency in camera synchronization and timestamp alignment.
- Experience with joint-space versus Cartesian command interfaces.
- Ability to identify issues in teleop demonstrations such as inconsistent grasps, occlusions, and timing skew.
- Experience in delivering projects with a written guide, video, and reproducible repository.
- Willingness to report n, the evaluation protocol, and failure modes publicly.
Skills
- Robot Learning
- Imitation Learning
- Policy Deployment
- Data Collection
- Hardware Integration
- Research Replication
- MLOps
- Robotics
- Computer Vision
- Software Engineering
Location
- San Francisco, CA
- Hong Kong
- Taipei
Work Type
- Full-time
- On-site preferred
Experience Level
- 2-4 years of fast-growing experience
- 1-2 years on a steep curve
Education Level
- Master's or Bachelor's in CS, robotics, or a related field
Benefits
- Health and Wellness
- Compensation and Support
About the Company
- Anvil is building the platform layer for Physical AI, creating accessible robotics hardware and software.
- The company has shipped over 200 robots and generated over $2.1M in revenue within its first year.
- Anvil designs and manufactures its own data collection devices in-house.
- The company has access to its own manufacturing facility and partner plants for diverse data collection.