We're partnered with a well-funded early-stage AI infrastructure startup building the low-latency inference platform that powers the next generation of intelligent machines.
This isn't a robotics research role.
You'll be responsible for taking cutting-edge AI models and making them run reliably in the real world, where milliseconds matter and model outputs directly control physical systems.
As one of the earliest engineers in the robotics team, you'll work across machine learning, distributed systems, networking, GPU inference, and robotics infrastructure to build the software layer between AI models and autonomous machines.
What you'll build
• Production inference systems serving robot policies under strict real-time latency requirements
• Low-latency infrastructure connecting AI models to robotic platforms
• High-performance integrations with robotics middleware and communication stacks
• GPU serving infrastructure for large vision-language and robotics models
• Evaluation systems for world models and autonomous policies
• Production tooling used directly by robotics customers deploying real-world AI systems
You'll likely have experience with
• Running ML models or learned policies on real robotic or autonomous systems
• GPU inference, model serving, and latency optimization
• Robotics middleware including ROS/ROS2
• Modern robot learning including VLAs, imitation learning, reinforcement learning, or world models
• Performance engineering across distributed systems, networking, and production infrastructure
• C++ and Python within large production codebases
• Debugging complex systems spanning perception, inference, networking, and hardware
Nice to have
• World models, diffusion models, or generative AI
• Teleoperation or real-time control systems
• TensorRT, CUDA, Triton, vLLM, SGLang, or other optimized inference frameworks
• WebRTC, QUIC, RTP, or other real-time networking technologies
• Linux performance engineering, GPU profiling, or systems optimization
• Multi-tenant GPU infrastructure or MLOps platforms
• Early-stage startup or founding engineer experience
Why join
• Build foundational infrastructure powering the next generation of physical AI
• Solve some of the hardest engineering problems at the intersection of AI, robotics, and distributed systems
• Work directly with leading robotics companies deploying production AI
• High ownership with significant influence over technical direction and architecture
• Competitive compensation with meaningful early-stage equity