Machine Learning Engineer, AI Inference Solutions (University Grad)

General Motors•Sunnyvale, CA

1d•Hybrid

About The Position

General Motors is a global leader in advanced driver assistance, with Super Cruise hands-free technology in more than 500,000 equipped vehicles on the road and over 700 million hands-free miles driven—demonstrating that automation can be trusted, intuitive, and helpful while reaching everyday drivers at unprecedented scale. Within GM AV, the Model Deployment & Inference Solutions team deploys machine learning models from training frameworks (e.g., PyTorch) onto autonomous-vehicle hardware; our two-fold mission is to build the ML deployment platform that makes model rollouts fast and predictable, and to optimize models so they meet the real-time latency and memory budgets required to run on-vehicle. Our work sits on the critical path for GM’s publicly committed launch of eyes-off (hands-free, eyes-free) autonomous driving in 2028 on the Cadillac Escalade IQ, and we’re hiring engineers to help deliver the next generation of safe, delightful personal autonomous-vehicle experiences. As an early career Engineer on the Model Deployment & Inference Solutions team, you’ll contribute across both sides of our mission: building the ML deployment platform and optimizing models for on-vehicle inference. You’ll work with and learn from senior engineers on real production deployments, platform features, and model-optimization workflows that ship to GM’s Super Cruise fleet at large scale, with structured mentorship and a clear onboarding plan. You’ll also collaborate closely with our sister teams (kernels, compiler, reduced precision, and parity) on the end-to-end path that takes trained models from research frameworks to ultra-efficient, safety-critical inference on the car. This is an early-career / new graduate role designed for candidates who have recently or will be completing their degree by June 2026.

Requirements

Recently completed or completing a Bachelor’s or Master’s degree by Spring 2026 in Computer Science, ECE, or a related technical field. (Degree must be completed before your start date.)
Strong computer science fundamentals (e.g., data structures, algorithms, operating systems, computer architecture) and solid coding skills in Python and/or C++, demonstrated through coursework, internships, or substantial projects.
Hands-on experience in AI/ML (e.g., machine learning, deep learning, computer vision, NLP, or ML systems) via classes, research, internships, or personal projects.
Depth in at least one of: computer architecture, operating systems, distributed systems, or compilers.
Demonstrated software-engineering experience (internships, coursework, open-source, research code, or competitions) showing good judgment around reliability, correctness, and clean abstractions.
Experience with—or strong interest in—using coding assistants/agents (e.g., Cursor, Claude Code, GitHub Copilot) as part of your workflow.
Ability to work effectively in collaborative, cross-functional teams and communicate clearly—both in writing and verbally—including explaining technical work partners

Nice To Haves

Internship, research, or advanced coursework in ML systems, ML compilers, GPU programming (CUDA, OpenAI Triton), inference optimization, or distributed training/serving infrastructure.
Familiarity with PyTorch and modern ML compiler/runtime stacks (e.g., torch.compile, TensorRT, ONNX, Triton Inference Server, vLLM, or equivalent).
Exposure to model optimization (quantization, pruning, distillation) or GPU profiling tools (Nsight Systems, Nsight Compute, PyTorch Profiler).
Familiarity with workflow/ML platforms such as Airflow, Temporal, Flyte, Ray, or Kubeflow.
Experience building agentic or LLM-powered tools or workflows.
Open-source contributions related to PyTorch, TensorRT, vLLM, OpenAI Triton, or similar projects.
Coursework, projects, or publications touching ML systems (e.g., MLSys, OSDI, ASPLOS, HPCA, NeurIPS systems track).
Familiarity with a systems language (e.g., C++) and development in a Linux environment.

Responsibilities

Contribute production code across the ML deployment platform, model-optimization workflows, and inference benchmarking/profiling infrastructure.
Pair with senior engineers on deployment workflows, performance investigations, model-optimization experiments (e.g., quantization, pruning, distillation), and platform tooling.
Build, test, and maintain platform tools (e.g., validators, performance probes, parity and sensitivity analyzers, agentic specialists) with technical guidance and code review support.
Investigate and help root-cause production deployment or performance issues; learn and apply the diagnostic playbook for compiler, kernel, runtime, and parity bugs.
Collaborate with cross-functional teams across the AV organization; including kernels, compiler, reduced-precision, parity, and model-development groups—to plan and execute model deployments to the AV stack, working under the guidance of senior engineers
Participate in code reviews, design discussions, and technical documentation to ensure reliability, correctness, and clear abstractions in a large-scale codebase.
Learn and follow secure coding, safety, and compliance practices required for on-vehicle autonomous driving software.