Internship 2027 Onboard Infrastructure Engineer, ML Inference
Core
Integrate LLMs and Vision-Language-Action (VLA) models into the onboard Rust middleware stack for autonomous heavy machinery, ensuring deterministic execution on edge compute.
Role type
Intern, Onboard Infrastructure Engineer (ML Inference)
Builds
Core engine for Bedrock's autonomous heavy machinery with real-time control and safety
Domain
Autonomous robotics, heavy construction equipment, edge AI
Deliverable
production ML models
Required skills
Rust, C++, PyTorch, JAX, GPU architectures, CUDA, parallel computing, multithreading, OS scheduling, memory management, asynchronous programming, IPC
Preferred skills
TensorRT, ONNXRuntime, ExecuTorch, KV-cache management, FP8/INT4 quantization, continuous batching, speculative decoding, multi-modal/VLA models, robotics frameworks
Technologies
Rust, C++, PyTorch, JAX, TensorRT, vLLM, ExecuTorch, NVIDIA Jetson Thor, Nsight Systems, Nsight Compute, eBPF
Responsibilities
Integrate open-source and proprietary LLM/VLA models into the onboard Rust middleware stack; Profile and optimize model execution using TensorRT, vLLM, or custom edge inference runtimes; Streamline sensor tokenization for real-time streams; Identify and eliminate bottlenecks in memory bandwidth, compute, and IPC; Validate performance optimizations on heavy autonomous machinery at test sites
Seniority
Intern
