Onboard Infrastructure Intern at Bedrock Robotics

Bedrock Robotics is hiring an Onboard Infrastructure Intern to integrate LLM and Vision-Language-Action models into its onboard Rust middleware and optimize their execution on edge hardware. The role involves profiling and tuning model inference with TensorRT, vLLM, or ExecuTorch on NVIDIA Jetson Thor, optimizing sensor tokenization, reducing memory, compute, and IPC bottlenecks, and validating performance on autonomous heavy machinery at test sites. Candidates must be pursuing a relevant degree and have Rust or C++ proficiency, experience with PyTorch or JAX, and knowledge of GPU architectures, CUDA, or parallel computing.