Job Description

Torc is seeking an engineer to optimize deep learning models for their virtual driver software used in autonomous trucks. The role involves developing custom CUDA kernels and TensorRT plugins, collaborating with domain experts, and reducing system performance bottlenecks.

Responsibilities

• Optimize deep learning models for embedded NVIDIA GPUs
• Develop custom CUDA kernels and TensorRT plugins
• Collaborate with domain experts
• Identify and reduce system performance bottlenecks
• Document work and participate in strategic discussions

Requirements

• Bachelor's degree with 5+ years of experience or advanced degree
• Proficiency with C++14
• Experience with CUDA programming and parallelization
• Experience writing TensorRT plugins
• Comfortable with collaborative development tools
• Strong written and verbal communication skills

Benefits

• No benefits