Manager, Engineering - Hardware Acceleration; CUDA

Manager, Engineering - Hardware Acceleration (CUDA)

At Torc, we have always believed that autonomous vehicle technology will transform how we travel, move freight, and do business. A leader in autonomous driving since 2007, Torc has spent over a decade commercializing our solutions with experienced partners. Now a part of the Daimler family, we are focused solely on developing software for automated trucks to transform how the world moves freight.

The Hardware Acceleration team builds the embedded inference foundation that enables Torc’s deep learning models to run efficiently and reliably on production AV hardware. Our mission is to convert and optimize perception and planning models for NVIDIA-based embedded platforms, develop custom CUDA kernels and pre/post-processing pipelines, and deliver deterministic, high-performance inference capabilities for the autonomy stack. We work closely with Perception, Application Engine, Systems, and Hardware teams to ensure that optimized models integrate seamlessly into Torc’s real-time autonomy platform.

We are seeking an experienced and technically strong Engineering Manager to lead Torc’s Hardware Acceleration group. This leader will shape the technical roadmap for model optimization and embedded inference while growing and guiding a high-performing team of engineers. Success requires both deep technical knowledge across CUDA, Tensor

RT, and embedded ML deployment, and the leadership ability to coach, hire, set direction, and drive execution in a safety‑critical environment.

• Lead the development and optimization of ML inference pipelines on embedded hardware, including model conversion (PyTorch/ONNX), Tensor
  
  RT integration, and CUDA‑based pre/post‑processing.
• Design, review, and guide development of custom CUDA kernels to support proprietary model layers and performance‑critical operations.
• Drive technical execution across model optimization, inference scalability, benchmarking, and real‑time system integration.
• Ensure high‑quality C++ and CUDA code through robust design, documentation, and test coverage.
• Integrate optimized models and processing stages into Torc’s Application Engine and support Virtual Driver teams in adopting the optimized inference layer.
• Hire, lead, and develop a high‑performance engineering team, building a culture of ownership, collaboration, and continuous improvement.
• Set technical and operational goals aligned with company‑wide objectives; define and track team KPIs and milestones.
• Provide coaching, career development, and skill‑building opportunities for engineers; maintain development plans and performance expectations.
• Establish and improve engineering processes for planning, delivery, testing, documentation, and cross‑team collaboration.
• Reinforce Torc’s values through transparent communication, conflict resolution, and proactive change leadership.

• Master’s degree in Computer Science, Electrical Engineering, or related field.
• Experience leading software engineering teams (people management, hiring, coaching, performance management).
• Strong technical expertise in CUDA, Tensor
  
  RT, NVIDIA Drive
  
  OS, and embedded inference workflows.
• Deep proficiency in C++ and modern software development practices.
• Experience with ML frameworks such as PyTorch and ONNX for model export and optimization.
• Strong Linux development experience and familiarity with real‑time, resource‑constrained systems.
• Background working with safety‑critical, automotive, or regulated environments.
• Ability to guide technical design decisions and challenge assumptions while fostering collaborative problem‑solving.
• Comfortable working in an agile, fast‑paced environment with shifting priorities.

• Hands‑on mentality — willing to jump into technical work when needed.
• Experience developing or certifying automotive‑grade products (ISO 26262, ASPICE).
• Experience building middleware, model‑serving frameworks, or GPU‑accelerated systems.
• Background in high‑performance computing, large‑scale model optimization, or distributed inference systems.

• A competitive compensation package that…