Jobs at Physical Intelligence

Own and optimize infrastructure for large-scale ML training and inference. Build JAX-based pipelines, manage GPU and TPU workloads, improve distributed performance, and create systems for scheduling, monitoring, checkpointing, debugging, and reproducible experimentation. Partner with researchers to translate modeling ideas into production training runs and contribute to core training code supporting new architectures, modalities, and evaluation metrics.

This role involves executing cross-functional projects, driving partnerships, managing deployment efforts, and supporting communications with investors and external parties.

Design and implement control algorithms for robotics, develop real-time control loops, and debug complex hardware/software issues while collaborating with cross-functional teams.

As a Forward Deployed Robotics Engineer, you'll manage robotic systems' technical issues in real-world settings, performing diagnostics, collaborating with teams, and analyzing data for resolution and improvements.

Deploy, debug, and iterate learned policies on physical robots across perception, policy, control, and hardware. Train and tune policies, curate data, build production-quality code, set up tests and data pipelines with operators, engage partners in deployment contexts, and translate research advances into reliable, real-world robotic systems.

Design and implement low-latency teleoperation interfaces and real-time video/streaming systems for on-site robots. Profile and optimize across drivers, networking, compute, and storage; build reliability, tracing, and debugging tooling; collaborate with researchers, hardware engineers, and operators to deploy dependable production robotics systems.

Engineer low-latency, high-throughput sensor and actuator pipelines, optimize system performance, and integrate complex models into real-time workflows while collaborating across teams.

Build internal production software for research and operations teams, starting with annotation tooling and expanding into researcher planning workflows, dataset browsing, evaluation dashboards, and operational tracking. Own requirements, prioritization, specifications, implementation, deployment, adoption, and iteration. Develop reliable frontend interfaces, backend APIs, data models, dashboards, and cloud services supporting workflows involving people, AI models, and physical operations.

Full-lifecycle technical recruiter owning searches for ML, robotics, systems, and infrastructure roles. Source passive top-tier candidates, design interview processes, advise on compensation and closing strategies, partner with founders and hiring managers, build market maps and dashboards, and steward candidate experience with data-driven sourcing and hiring approaches.

Own quality assurance and auditing for multimodal data sourced from external vendors. Build scalable software tooling and infrastructure for sampling, review workflows, quality checks, audit trails, and vendor reporting. Define measurable data standards with researchers, manage vendor feedback loops, and improve data quality through instrumentation and iterative processes. The role also requires hands-on sample review, rubric development, cross-functional communication, and approximately five years of customer-facing software experience.

Lead and scale data annotation operations from hundreds to thousands, owning throughput, quality, cost, hiring, training, tooling, vendor mix, and budget. Build multi-layer management, human-in-the-loop/autolabeling workflows, QA/rubrics, operational metrics, capacity planning, and partner with product, engineering, and research to meet SLAs and improve unit economics.

Build and operate large-scale data infrastructure for robot learning: ingest, validate, transform, and featurize multimodal data; run batch and streaming workflows; design storage and metadata systems; optimize training-time performance; manage data movement and lifecycle; implement observability and guardrails; collaborate with researchers and engineers.

Own product roadmap for Research Operations; design and improve systems for intake, prioritization, resource allocation, execution visibility, quality management, and operational planning. Partner with engineers, project leads, Production Operations, and researchers to build tools, standardize processes, establish metrics and dashboards, remove bottlenecks, and prioritize investments that scale research execution.

The Recruiting Coordinator will schedule interviews, manage candidate pipelines, enhance recruiting workflows, support hiring managers, and maintain candidate communication.

Own delivery of company-critical research and deployment initiatives: define scope, prioritize work, allocate resources, build execution plans, direct cross-functional execution, resolve risks and bottlenecks, run stakeholder communications, and improve operational systems as the organization scales.

Manage HR operations including onboarding, payroll administration in Rippling, benefits and HRIS data integrity, compliance with Federal/CA/SF laws, and partner on employee relations, leaves, and HR process design to support growth.

This position offers internship opportunities for individuals interested in gaining research experience.

Lead end-to-end NPI program execution for robotic hardware platforms: schedule and risk management, material forecasting, CTB analysis, build planning, vendor coordination, logistics readiness, configuration management, and cross-functional handoffs to fleet integration to support research and deployments.

As an ML Infra Engineer, you will optimize and manage training systems for large-scale models, collaborating with researchers and engineers on infrastructure and experiment scalability.

The Build & Release Engineer will manage infrastructure and developer platforms, focusing on CI/CD systems, build reliability, and Kubernetes services to improve developer productivity and operational efficiency.