Robotics Engineer

OpenMind; San Francisco (CA); On-site; USD 120,000 - 190,000; Full time

Job summary

OpenMind in San Francisco is seeking an experienced robotics software engineer to own core autonomy capabilities across perception, SLAM, navigation and manipulation. You will bring up new hardware platforms from unboxing to deployment-ready, and ship software that runs on real robots in customer environments.

You should have hands-on experience with ROS2, Python/C++, and embedded compute platforms (e.g., NVIDIA Jetson).

Qualifications

• Hands-on depth in at least two of SLAM, motion planning, perception, manipulation, sensor calibration.
• Experience with ROS2 and embedded compute platforms.
• Shipped software on real hardware outside a lab.
• Comfort owning problems across the stack, from driver-level issues to high-level behavior.

Responsibilities

• Develop and harden core autonomy: 3D SLAM and navigation, localization, dynamic obstacle handling, and multi-floor traversal.
• Bring up new robot platforms: sensor integration, calibration, firmware interfaces, and SDK integration across vendors.
• Build shared fleet capabilities such as common memory and map sharing so that mapping one robot means mapping the fleet.
• Work on real-time performance: perception latency, compute budgeting on embedded platforms (NVIDIA Jetson and Thor), and sensor pipelines.
• Support deployments by making the software robust to the messiness of real environments: slopes, crowds, poor lighting, and unreliable networks.

Skills

Robotics software

Python

C++

Go

ROS2

Embedded compute

SLAM

Perception

Navigation

Tools

ROS2

NVIDIA Jetson

Embedded compute platforms

Job description

The Role

You will make robots work. OpenMind supports multiple platforms today, from quadrupeds to full-size humanoids, and now our own robots are here. You will own core autonomy capabilities across perception, SLAM, navigation, and manipulation, and you will bring up new hardware platforms from unboxing to deployment-ready.This is a role for an engineer who wants their code running on physical robots in customer environments, not in a simulator queue.

What You Will Do

• Develop and harden core autonomy: 3D SLAM and navigation, localization, dynamic obstacle handling, and multi-floor traversal
• Bring up new robot platforms: sensor integration, calibration, firmware interfaces, and SDK integration across vendors
• Build shared fleet capabilities such as common memory and map sharing so that mapping one robot means mapping the fleet
• Work on real-time performance: perception latency, compute budgeting on embedded platforms (NVIDIA Jetson and Thor), and sensor pipelines
• Support deployments by making the software robust to the messiness of real environments: slopes, crowds, poor lighting, and unreliable networks

What We Look For

• Experience building software for physical robots, with strong engineering fundamentals in languages such as Python, Go, and/or C++
• Hands-on depth in at least two of: SLAM, motion planning, perception, manipulation, sensor calibration
• Experience with ROS2 and embedded compute platforms
• Comfort owning problems across the stack, from driver-level issues to high-level behavior
• Shipped software running on real hardware outside a lab

Nice to Have

• Experience with humanoids, quadrupeds, or other form factors such as wheeled robots
• Experience building on top of pretrained vision-language-action foundation models and adapting them to new embodiments and tasks
• Experience applying reinforcement learning to robot control problems such as locomotion, whole-body control, or manipulation, using frameworks like Isaac Lab or MJLab
• Familiarity with NVIDIA Isaac, GMSL camera pipelines, or real-time middleware
• Contributions to open-source robotics software