Torc Robotics
Staff, ML Engineer - Scene Generation
$250K–$300K
Remote - US, Ann Arbor, MI +2 · Remote · Staff · Full-time · Posted Aug 4, 2026
- Location
- Remote - US, Ann Arbor, MI
Ann Arbor, Michigan, United States
Remote - US - Experience
- The posting gives education-linked alternatives: Bachelor’s plus 10+ years, Master’s plus 7+ years, or PhD plus 5+ years. It also separately asks for a PhD or equivalent work experience of 8+ years in a relevant field.
- Education
- Bachelor’s, Master’s, or PhD in Computer Science, Robotics, Electrical Engineering, or a related technical field; the posting also says “PhD or equivalent work experience.”
About the job
Leads the Scene Generation team in developing and deploying neural-rendering and generative-model sensor simulation for autonomous-truck training and validation.
What you'll do
- Set technical vision and roadmap, lead day-to-day priorities, and make architecture decisions.
- Guide research and production implementation of camera, LiDAR, and Radar simulation; own the framework through testing, cloud integration, and deployment.
- Align with other teams and stakeholders, mentor engineers, and support hiring and career development.
What they ask for
- Proficiency in Python and deep learning frameworks such as PyTorch.
- Expertise in Neural Rendering (Neural Radiance Fields and 3D Gaussian Splatting) and generative models (Diffusion Models, Flow Matching).
- Production software experience and technical team leadership, including setting direction and driving consensus.
- Experience with production robotics and/or autonomous-driving data; background in Computer Vision, Computer Graphics, 3D Reconstruction, or 3D Computer Vision.
Nice to have
- Proficiency with CUDA programming for efficient rendering of large-scale scenes.
- Experience with MLOps and infrastructure tools such as Ray.
Pay
$249,600 — $299,500 USD (total compensation package)
Tools
Python, PyTorch, Neural Radiance Fields, 3D Gaussian Splatting, Diffusion Models, Flow Matching, CUDA, Ray
AI focus
Neural rendering and generative modeling for camera, LiDAR, and Radar sensor simulation, supporting autonomous-driving training and validation.
Model-written summary · the original posting is authoritative