Director of Engineering, Physical AI

Scale AI Scale AI · Data AI · San Francisco, CA · Physical AI Engineering

Director of Engineering for Physical AI at Scale AI, leading the execution of the Physical AI Data Engine. Responsibilities include setting technical vision for data collection, teleoperation, ML training, evaluation, and annotation tooling, while managing a multidisciplinary team and ensuring cross-functional alignment for robotics workloads.

What you'd actually do

  1. Set and drive the technical vision across data collection infrastructure, teleoperation systems, ML training pipelines, model evaluation frameworks, annotation tooling, and research
  2. Lead a multidisciplinary engineering organization—spanning engineering managers, software engineers, ML engineers, and ML research scientists—while designing the organizational structure, talent strategy, and culture required to scale rapidly without compromising on quality or strategic alignment
  3. Maintain exceptional technical and operational excellence by deeply understanding team deliverables, asking incisive questions, identifying slipping standards early, and knowing precisely when to step in
  4. Drive cross-functional alignment across Engineering, Operations, and GTM on platform architecture, release processes, and shared priorities
  5. Collaborate with researchers and clients to architect and deliver scalable, production-grade data infrastructure tailored for complex robotics workloads

Skills

Required

  • Bachelor's degree in Engineering, Robotics, Computer Science, or a related technical field
  • 8+ years of engineering experience in fast-paced environments
  • 4+ years direct people management demonstrated history of recruiting, mentoring, and developing high-performing technical teams
  • Experience leading technical execution for complex hardware-software systems
  • Deep domain knowledge in Robotics, Autonomous Vehicles, Computer Vision, and/or Machine Learning
  • Deep fluency in the ML development lifecycle — training pipelines, data flywheels, and evaluation frameworks
  • Leading teams across Python, C++, and TypeScript/Node stacks
  • Distributed systems
  • Cloud infrastructure (AWS, Kubernetes)
  • Workflow orchestration (Temporal, Airflow)
  • Proven ability to independently navigate, execute effectively amidst ambiguity
  • Strong attention to detail
  • Strong operator and communicator

Nice to have

  • MS or PhD
  • Hands-on experience with teleoperation systems (ALOHA, UMI, hand tracking)
  • Robotic hardware platforms
  • Imitation learning
  • Sensor fusion
  • SLAM
  • 3D data processing
  • Experience scaling data collection systems globally

What the JD emphasized

  • own the execution of the Physical AI Data Engine
  • solve the data bottleneck that stands between today's robotics research and real-world deployment
  • Deep fluency in the ML development lifecycle — training pipelines, data flywheels, and evaluation frameworks as systems you've built and owned

Other signals

  • leading a multi-disciplinary engineering organization
  • own the execution of the Physical AI Data Engine
  • solve the data bottleneck that stands between today's robotics research and real-world deployment
  • ML training pipelines, model evaluation frameworks, annotation tooling
  • architect and deliver scalable, production-grade data infrastructure tailored for complex robotics workloads