AI SWE - Edge Model Optimization

Overview

Samson Rose has been exclusively engaged by a pioneering Robotics & AI company to find a Staff AI Software Engineer focused on optimizing and deploying deep learning models on edge hardware for real-world robotic systems. This is a senior, hands-on role centered on translating cutting-edge AI research into production-grade autonomy. You will work at the intersection of AI, robotics, and embedded systems, ensuring models run reliably under strict constraints on latency, power, and memory in challenging field environments.

A bit about the company: They build embodied AI systems that allow robots to perceive, reason, and act directly on-device, without reliance on cloud compute or curated environments. Their autonomy stack runs on embedded platforms such as NVIDIA Jetson and Orin, enabling continuous operation in harsh, real-world conditions. With substantial funding, the company supports long-term research, deployment, and iteration  team includes experts from Deep Mind, NASA JPL, Boston Dynamics, NVIDIA, Amazon, Tesla Autopilot, Cruise, Zoox, Toyota Research Institute, and Space

X, with a decade-long history of successful field deployments and DARPA challenge wins.

• BS, MS, PhD, or equivalent experience in Computer Science, Robotics, Electrical or Computer Engineering, or a related field.
• 5+ years of professional experience developing and deploying deep learning models for edge, embedded, or real-time systems.
• Strong proficiency in PyTorch, C++, Python, and CUDA.
• Hands-on experience with Tensor
  
  RT, ONNX, and Triton, including custom Tensor
  
  RT plugin development.
• Proven experience applying model optimization techniques such as quantization, pruning, and distillation in production systems.
• Deep understanding of performance tuning on Jetson or ARM platforms, GPUs, and embedded Linux.
• Experience integrating AI models into ROS-based robotic systems.
• Ability to work independently while collaborating effectively across AI, robotics, and hardware teams.

• Model Optimization & Deployment
• Convert and optimize 2D and 3D CNNs and Transformer-based models for real-time inference on Jetson and Orin platforms using ONNX, Tensor
  
  RT, and Triton.
• Performance Engineering
• Apply compression techniques and develop custom Tensor
  
  RT plugins and CUDA kernels to meet strict latency, memory, bandwidth, and power constraints.
• System Integration & Validation
• Integrate optimized models into ROS-based robotic systems, build benchmarks, profile end-to-end pipelines, and validate performance in real-world robotic deployments.

If this role is of interest to you, please apply with your current resume. We will reach out to schedule an initial call.