29 days ago
Milpitas, CA, USASenior / Staff+
H1B sponsor
Base Salary
$185k - $325k/yr
Responsibilities
- Design and maintain multimodal data-collection pipelines for teleoperation devices and UMI hardware, including synchronization, versioning, and distributed storage.
- Build annotation tooling and data-curation workflows for quality filtering, deduplication, episode scoring, and domain reweighting.
- Develop post-SFT reinforcement-learning infrastructure for demonstration reward scoring, failure-pattern mining, and retraining feedback.
- Build evaluation and test infrastructure to log on-robot policy rollouts, capture structured results, and provide diagnostics.
- Collaborate with ML researchers on data schemas, episode formats, and pipeline interfaces for VLA and manipulation-policy training.
- Architect scalable cloud and on-premise storage and retrieval systems for heterogeneous robot data, including vision, proprioception, action, and language data.
Requirements
- Bachelor’s or master’s degree in Computer Science, Robotics, or a related field with 5+ years of experience.
- Strong proficiency in Python and experience building production-grade data pipelines and ETL systems.
- Hands-on experience with large-scale dataset management, versioning, deduplication, quality filtering, and distributed storage.
- Experience with post-training infrastructure such as SFT pipelines, reward modeling, or reinforcement-learning training loops.
- Familiarity with PyTorch, JAX, and ML training workflows sufficient for close collaboration with research teams.
- Bonus: experience with teleoperation, UMI, GELLO, or similar robotics data-collection hardware.
- Bonus: familiarity with imitation learning, behavior cloning, VLA/VLM fine-tuning, evaluation infrastructure, experiment tracking, rollout logging, or human-in-the-loop annotation systems.
Benefits
- Competitive stock options and equity programs.
- Health, dental, and vision insurance and a 401(k) plan.
- Visa sponsorship and green card support for qualified candidates.
- Lunches and dinners, a fully stocked kitchen, and regular team-building events.
- Full-time U.S. position requiring five days per week of in-office collaboration.
Categories
Data EngineeringML Engineering
