Software Engineer II, MLOps Framework

About this role

Torc is seeking a Software Engineer II for MLOps Framework to own the end-to-end pipeline that takes machine learning models from research to production deployment on autonomous truck edge hardware. You'll architect model conversion pipelines, maintain release registries, and validate that deployed models meet strict latency and accuracy requirements for safety-critical autonomous driving applications.

What you'll do

• Architect and implement model conversion and compilation pipelines using ONNX, TensorRT, and torch.compile for NVIDIA Orin and similar edge devices
• Maintain and evolve the model release registry to ensure traceability and reproducibility across versions and platforms
• Perform rigorous latency benchmarking and model quality parity validation against safety-critical requirements
• Compare performance metrics across platforms to verify accuracy and latency compliance before release
• Collaborate with perception, safety, and model development teams to translate technical findings into actionable outcomes

What they're looking for

• Model conversion and compilation pipelines (ONNX, TensorRT, torch.compile)
• Latency benchmarking and performance validation
• Edge hardware deployment (NVIDIA Orin and embedded platforms)
• Model release registry management in production environments
• Performance metrics analysis and interpretation
• Model quantization (PTQ, QAT) and mixed-precision inference
• Autonomous driving perception algorithms
• C++ and/or CUDA kernel development

Benefits

• Remote work opportunity across the US
• Collaborative and energetic team culture
• Health and work/life balance support
• Opportunity to work on pioneering autonomous vehicle technology
• Partnership with Daimler and industry-leading partners