> Markdown version of [/jobs/ext/4667552-staff-software-engineer-imitation-learning-behavior-planning-embodied-ai](https://www.wearedevelopers.com/jobs/ext/4667552-staff-software-engineer-imitation-learning-behavior-planning-embodied-ai). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Staff Software Engineer, Imitation Learning - Behavior Planning, Embodied AI - **Company:** General Motors - **Location:** Sunnyvale, CA, United States - **Salary:** $189,000 - $280,000 - **Contract:** Permanent contract - **Skills:** Data Cleansing, Extract Transform Load (ETL), Distributed Computing Environment, Python (Programming Language), Machine Learning, Reinforcement Learning, Data Processing, Pytorch, Embodied AI, Information Technology, Codebase, Machine Learning Operations, TensorRT, Data Generation - **Published:** November 11, 2026 - **Apply:** https://generalmotors.wd5.myworkdayjobs.com/Careers_GM/job/Sunnyvale-California-United-States-of-America/Staff-Software-Engineer--Imitation-Learning---Behavior-Planning--Embodied-AI_JR-202623706 ## About the Role * Master's or PhD degree in Computer Science, a related field, or equivalent practical experience. * Track record of leading complex software projects, setting architecture, and delivering hands-on implementation across team boundaries. * Strong Python skills and experience building maintainable, well-tested software in large codebases. * Hands-on experience with PyTorch or comparable frameworks and ML training or evaluation systems. * Experience profiling and optimizing performance-sensitive ML pipelines, including data processing, CPU/GPU execution, and distributed training. * Strong technical judgment and ability to translate research needs into reliable production software. Preferred: * Experience in autonomous driving, robotics, behavior planning, or trajectory prediction. * Familiarity with imitation learning, RL, multistage training, distillation, or synthetic data generation. * Experience with closed-loop simulation, model deployment, and training-inference consistency testing. * Modern C++ skills or experience with GPU profiling, model compilation, and ML runtime optimization. ## Description * Architect and build modular ML software spanning data preparation, training, evaluation, and production integration. * Optimize end-to-end training and evaluation performance, including data loading, CPU/GPU utilization, distributed execution, and workflow orchestration. * Enable reusable IL pretraining and rapid fine-tuning and post-training workflows, including integration with reinforcement learning (RL). * Integrate open-loop behavioral tests, closed-loop simulation, and targeted scenarios to provide fast, actionable feedback on model quality. * Build reusable model interfaces and data augmentation tools; maintain consistency across training, simulation, and on-vehicle inference. * Lead cross-team technical initiatives, mentor engineers, and improve software modularity, testing, reproducibility, and reliability.