Sr. Staff Engineer, Autonomy Triage

Rivian•Palo Alto, CA

2d•Onsite

About The Position

The Autonomy Triage team is responsible for the triage infrastructure and workflows that close the loop between what drivers experience on the road and how our autonomy stack evolves. As a Sr Staff Triage Engineer, you'll own the architecture of the systems and methodology that turns unlabeled miles into actionable signals; defining how we detect, classify, and root-cause undesired behavior at scale. This is a technically deep, high-ownership role at the intersection of data engineering, ML systems, and autonomy validation. You'll work directly with perception, planning, systems, and safety teams to ensure the right events surface from hundreds of millions of miles of customer and simulation data; and that the signatures we build today will be relevant and applicable as our system scales across both consumer features and L4 robotaxi.

Requirements

Graduate degree in engineering, preferably in CS, electrical or mechanical engineering
10+ years of engineering experience, with at least 5 years developing autonomous vehicle systems, preferably Level 4 systems.
Deep familiarity with the full AV development loop, autonomy systems, sensor modalities, and end-to-end triage process; data collection, labeling pipelines, metric frameworks, simulation, and safety validation
Hands-on experience designing or owning triage, disengagement analysis, or event detection systems at scale
Strong problem solving skills in a highly cross functional and fast paced environment
Strong data engineering fundamentals: large-scale data pipelines, feature extraction, and event querying across structured and unstructured driving data
Experience with ML model development or deployment — particularly classification or anomaly detection in time-series or sensor data
Proficiency with Python and SQL

Nice To Haves

experience working with LLMs or prompt-based pipelines is a strong plus

Responsibilities

Define and lead the strategy and execution for L4 autonomy triage at scale
Define and own the end-to-end triage pipeline: from event detection through feature metric parity analysis, classification, and root cause analysis
Enhance the ground truth infrastructure that ties driver-reported symptoms to event signatures, enabling precision and recall measurement of our feature metrics
Drive the methodology for evaluating metric parity between on-road and simulation environments
Ensure on-road signatures and sim metrics are in sync and representative of real world scenarios, and that gaps are continuously closed through focused feedback cycles
Establish coverage benchmarks and drive the roadmap for expanding triage coverage across new scenarios, geographies, and ODD boundaries