Physical Intelligence
Software Engineer, Data Quality
San Francisco · On-site · Full-time · Posted Aug 25, 2026
- Experience
- ~5 years across building customer facing software.
About the job
Owns quality assurance for third-party multimodal data, defining standards, building QA systems, and improving data quality with vendors before data is used for training.
What you'll do
- Build and run end-to-end QA and audit processes for vendor data, including sampling, review workflows, decisions, and audit trails.
- Work with researchers to define and maintain measurable data-quality standards and acceptance criteria.
- Build automated QA tools and infrastructure, and deliver feedback to vendors to improve their data quality.
What they ask for
- Experience building customer-facing software.
- Experience managing vendors, delivering actionable feedback, and measurably improving their output.
- Strong operations skills, including instrumenting, measuring, and iterating on processes.
- Hands-on experience reviewing data samples, writing rubrics, and checking program output.
Nice to have
- Experience with robotics, egocentric video, or other multimodal sensor data.
- Experience defining requirements for QA/validation tooling or working with engineers on automated data quality checks.
AI focus
Supports training robot foundation models by ensuring multimodal data is high quality and verified before training.
Model-written summary · the original posting is authoritative