Senior Project Manager, Nuro

• This role demands someone wired for that level of precision, someone for whom "close enough" is never acceptable and who can see the difference between a labeling error and a labeling system error
• 5+ years of project or program management experience embedded with ML, data operations, or software engineering teams. You have been close to the work, not managing from a distance
• Quality-obsessed, not just execution-obsessed. You are energized by finding what is broken, not just by shipping on time. You ask why accuracy is degrading before asking when the pipeline will recover
• Visionary and hands-on in equal measure. You can design the system-level quality framework and personally dig into a database to trace where a labeling fault originated. You do both without being asked
• Hands-on data fluency: you can navigate a database schema, write SQL to investigate a labeling anomaly, and form a diagnosis without waiting for a data engineer to pull the report for you
• Direct experience with or deep understanding of ML data pipelines and data labeling ecosystems, including annotation workflows, quality sampling methodologies, labeling taxonomy design, inter-annotator agreement, and the levers that drive or degrade training data accuracy
• Proven ability to identify systemic workflow problems across multiple concurrent pipelines, propose targeted improvements, and drive adoption across teams
• Approaches problems with intellectual curiosity and analytical rigor, taking the time to form a well-supported point of view and communicate it with confidence
• Exceptional communication: you translate a nuanced data quality finding into a precise business or safety risk that a senior executive can act on
• Bachelor's degree in a technical or business discipline, or equivalent practical experience

Desirable

• As a Senior Project Manager, ML Operations, you are not an execution manager. The team already has operators focused on keeping pipelines running
• What we need is someone who steps back, looks across all of them, and asks: where is quality breaking down, why, and what do we do about it? You will co-own more than 10 active data labeling pipelines alongside senior leadership, with a singular focus on data quality and gap closure
• You will be the person who identifies the signal in the noise, connects dots across pipelines that execution-focused OPMs cannot see from their vantage point, and drives the systemic improvements that raise the accuracy of our training data. Data labeling at this scale is unforgiving
• A single systematic flaw in how objects are annotated or how edge cases are classified can propagate silently through training and surface as a safety regression on the road
• Own the data quality standard across 10+ labeling pipelines. Establish what "good" looks like for each data type, instrument the pipelines to measure against it, and track gaps with rigor
• Identify and close quality gaps. Audit live workflows, query databases, trace accuracy failures to their structural root cause, and return with a specific, evidence-based improvement plan
• Operate above the execution layer. Partner with operations project managers who own throughput and delivery, and provide the quality diagnostic lens they are not positioned to hold
• Design and implement scalable processes to improve labeling accuracy, reduce systematic errors, and support evolving ML training requirements
• Serve as the strategic interface between Autonomy Engineering and ML Operations, connecting labeling quality metrics directly to model performance and safety outcomes
• Build executive-ready reporting that frames quality gaps and improvement progress as model performance and safety signals, not just operational metrics
• Drive cross-functional alignment across engineering, product, and global ops teams by bringing clear analysis and well-reasoned recommendations to every conversation