Staff Machine Learning Engineer- Large Driving Model (LDM), Motion Planning, Autonomy

Rivian is building the next generation of autonomy through a Large Driving Model (LDM) powering the Rivian Autonomy stack. As a Staff Machine Learning Engineer, you will lead the end-to-end lifecycle of the LDM, from data curation and post-training pipelines to evaluation and on-road production performance. This role combines deep model expertise with real-world autonomy impact, with technical leadership across the team and the autonomy roadmap.

What you’ll work on

• Spearhead development of advanced driving capabilities by owning the full LDM lifecycle, including targeted data curation, post-training alignment, and rigorous evaluation pipelines.
• Assess and improve model design across supervision quality, validation strategies, and scalable architectures that close the gap between offline LDM training and on-road performance.
• Serve as a technical authority within the Autonomy department, providing definitive guidance on technology decisions, model architectures, and data strategies.
• Guide tradeoffs with a strong understanding of business, safety, and operational implications, aligning decisions to company timelines and product needs.
• Enable other engineers by architecting scalable, extendable, and collaborative ML systems and codebases, and establishing best practices for the broader LDM team.
• Influence technical roadmaps and build consensus across perception, planning, and platform teams, including during challenging technical discussions.
• Deploy, supervise, and iteratively enhance models as they transition from development into customer vehicles, with a focus on verified real-world performance.

What you bring

• M.S. or Ph.D. in Machine Learning, Computer Science, Robotics, or a related field.
• 7+ years of industry experience building complex, production-grade ML systems, with experience operating at a Staff or Lead level.
• Expert-level, hands-on experience with Python and PyTorch.
• Large-scale distributed training frameworks experience is highly preferred.
• Deep expertise in foundation models and post-training methods (including fine-tuning and alignment), plus experience applying approaches for perception, prediction, or planning in autonomous driving or robotics.
• Proven ability to own ML data verticals, including data curation, synthetic data generation, and building robust evaluation frameworks for complex behavioral models.
• Solid fundamentals in linear algebra, geometry, optimization, algorithms, and data structures.
• Exceptional communication skills, with proven ability to set technical direction, guide execution, and build consensus in a fast-paced environment.
• Motivation to move ambiguous, complex ideas from development and validation through to a verified, shipping product.
• Hands-on mindset: proactively identify systemic gaps, write critical-path code, and address technical voids as needed.
• Prior experience with Automotive Self-Driving stacks or similarly complex, safety-critical robotics systems.

Location: Palo Alto, CA (onsite)

Technologies: Python, PyTorch