
Fullstack Software Engineer
Physical Intelligence15 days ago
Responsibilities
- Build annotation tooling and workflows, including annotator interfaces and supporting pipelines.
- Develop systems that track annotation quality, cost, throughput, and coverage.
- Redesign researcher request and planning workflows and build capacity and prioritization software.
- Create dataset browsing, evaluation dashboard, and Production Ops tooling for lab, warehouse, and real-world deployments.
- Gather requirements, prioritize work, define success metrics, write specifications, ship tools, drive adoption, and iterate from feedback.
- Build, ship, and support production-quality frontend interfaces, backend APIs, data models, dashboards, and cloud services.
Requirements
- Strong full-stack experience building production web applications and APIs, especially with React, TypeScript, and Python.
- Experience with relational databases including Postgres, analytical systems including ClickHouse, and queueing systems.
- Familiarity with cloud and containerized environments such as GCP and Kubernetes.
- Ability to design workflows integrating people, models, and software; exercise product judgment; and navigate ambiguous requirements.
- Experience working directly with users and iterating based on feedback.
- Ability to start with a practical initial version and develop it into scalable, production-quality systems.
- Preferred experience includes founding or joining an early-stage company, building data labeling or human-in-the-loop tools, developing internal tools for research or robotics, and working with tasking, instruction management, scheduling, resource allocation, or workflow orchestration systems.