Job description

JOIN THE TEAM BRINGING AUTONOMY TO THE BUILT WORLD At Bedrock, we're moving AI out of the lab and onto real job sites. Our team includes veterans who helped launch Waymo, scaled Segment to a $3.2B acquisition, and grew Uber Freight to $5B in revenue. We retrofit the heavy equipment contractors already own, starting with excavators, so experienced crews can safely take on more projects and deliver them more predictably. In August 2026, Bedrock-equipped excavators began doing fully autonomous, paid work on live commercial job sites. We've more than doubled our customer base since 2025, and our machines now work with contractors across the West, South, and Midwest. In October 2026, we raised a $600M Series C co-led by CapitalG and Emergence Capital, bringing our total funding to nearly $1B. We're using it to put more autonomous machines to work, build the field and operations teams behind every deployment, and expand into new machine types and job-site tasks. This is where algorithms meet steel-toed boots. You'll work alongside construction veterans and world-class engineers to solve physical-world problems that simulations can't touch. If you're ready to do meaningful work on hard problems, we'd love to have you join us. ABOUT THE ROLE & TEAM The Onboard Infrastructure team builds the core engine of Bedrock’s autonomous heavy machinery — where safety, real-time control, and millisecond-level responsiveness are non-negotiable. As an Onboard Infrastructure Intern, you will bridge frontier AI and real-time execution by integrating Large Language Models (LLMs) and Vision-Language-Action (VLA) models into our core stack, ensuring multi-billion parameter multi-modal models run within strict, deterministic deadline budgets on edge compute. WHAT YOU'LL DO - Integrate open-source and proprietary LLM/VLA models into our onboard Rust middleware stack alongside existing perception, planning and control pipelines. - Profile and optimize model execution using TensorRT, vLLM, ExecuTorch or custom edge inference runtimes tailored for NVIDIA Jetson Thor. - Streamline sensor tokenization (cameras, LiDAR) to feed real-time streams directly to models without latency spikes in vehicle control loops. - Identify and eliminate bottlenecks across memory bandwidth, compute, and IPC using tools like Nsight Systems, Nsight Compute and eBPF. - Validate your performance optimizations directly on heavy autonomous machinery at our test sites. WHAT WE'RE LOOKING FOR REQUIRED - Currently pursuing a BS, MS, or PhD in Computer Science, Electrical/Computer engineering, Robotics or a related field. - Proficiency in Rust or C++, with supporting experience in Pytorch or JAX. - Solid foundation in GPU architectures, CUDA, or parallel computing. - Understanding of modern systems concepts: multithreading, OS and GPU scheduling, memory management, asynchronous programming and IPC. BONUS POINTS - Practical experience deploying neural networks on constrained hardware using TensorRT, ONNXRuntime or ExecuTorch. - Experience with LLM/VLA optimization techniques such as KV-cache management, FP8/INT4 quantization, continuous batching or speculative decoding. - Exposure to multi-modal/VLA models or robotics frameworks. About This Internship This is a fixed-term, project-based internship designed strictly for students (undergraduate or graduate) seeking resume-building experience, hands-on skill development, or industry exposure rather than a direct career launchpad. Candidates will own discrete, short-term projects that can be fully completed within the internship timeframe. Please note that while exceptional interns may occasionally be considered if business needs align, opportunities for full-time conversion are highly limited." Bedrock Robotics is an Equal Opportunity Employer We’re committed to building a diverse and inclusive workplace. We consider all qualified applicants for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, ancestry, age, disability, veteran status, genetic information, or any other protected characteristic. Reasonable Accommodations We want our hiring process to be accessible to everyone. If you need an accommodation to participate in the application or interview process, please let your recruiter know so we can support you.

Requirements

Must-haves

JOIN THE TEAM BRINGING AUTONOMY TO THE BUILT WORLD
Integrate open-source and proprietary LLM/VLA models into our onboard Rust middleware stack alongside existing perception, planning and control pipelines.
Profile and optimize model execution using TensorRT, vLLM, ExecuTorch or custom edge inference runtimes tailored for NVIDIA Jetson Thor.
Streamline sensor tokenization (cameras, LiDAR) to feed real-time streams directly to models without latency spikes in vehicle control loops.
Identify and eliminate bottlenecks across memory bandwidth, compute, and IPC using tools like Nsight Systems, Nsight Compute and eBPF.
Validate your performance optimizations directly on heavy autonomous machinery at our test sites.
Currently pursuing a BS, MS, or PhD in Computer Science, Electrical/Computer engineering, Robotics or a related field.
Proficiency in Rust or C++, with supporting experience in Pytorch or JAX.
Solid foundation in GPU architectures, CUDA, or parallel computing.
Understanding of modern systems concepts: multithreading, OS and GPU scheduling, memory management, asynchronous programming and IPC.

Nice-to-haves

Practical experience deploying neural networks on constrained hardware using TensorRT, ONNXRuntime or ExecuTorch.
Experience with LLM/VLA optimization techniques such as KV-cache management, FP8/INT4 quantization, continuous batching or speculative decoding.
Exposure to multi-modal/VLA models or robotics frameworks.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aillmrust

Platform: ashby
Effort: quick

Benefits

Benefits detected in the listing.

Vision