ML API Platform Engineer for Scalable Robotics

Physical Intelligence; San Francisco (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Physical Intelligence in San Francisco seeks an engineer to build the model API end-to-end, from data ingestion to low-latency inference, for external partners.

You'll work with researchers to turn new model capabilities into stable surfaces, design scalable, secure multi-tenant systems, and empower developers with tools and documentation.

This role emphasizes ownership, production-grade code, and delivering reliable, scalable robotics software.

Qualifications

• Experience deploying ML systems to production environments.
• Ability to design scalable developer platforms.
• Familiarity with security and multi-tenant infrastructure.
• Experience with model fine-tuning or inference workloads.
• Ability to write high-quality production code.

Responsibilities

• Build Pi’s model API end to end with data ingestion and fine-tuning.
• Design scalable multi-tenant infrastructure with observability and SLOs.
• Collaborate with researchers and engineers to turn model capabilities into APIs.
• Develop partner-facing tooling and deployment integrations.
• Write production-grade code with deep integration to existing infra.

Skills

Python

Distributed systems

Production systems

Model serving

DevOps / infrastructure

Ownership

APIs

Security / multi-tenant

Full-stack (React/TS)

Tools

Kubernetes

PostgreSQL

ClickHouse

GCP

TypeScript

React

Job description

Physical Intelligence in San Francisco seeks an engineer to build the model API end-to-end, from data ingestion to low-latency inference, for external partners.

You'll work with researchers to turn new model capabilities into stable surfaces, design scalable, secure multi-tenant systems, and empower developers with tools and documentation.

This role emphasizes ownership, production-grade code, and delivering reliable, scalable robotics software.