Aurora Innovation
Senior Software Engineer, ML Evaluation & Introspection Infrastructure
$146K–$260K
Seattle, WA +2 · Hybrid · Senior · Full-time · Posted Oct 19, 2026
- Location
- Seattle, WA
Mountain View, CA
Pittsburgh, PA - Experience
- Typically 5+ years of professional software engineering experience building backend, data, or ML infrastructure.
About the job
Build and own infrastructure for evaluating, searching, and explaining self-driving models using real-world driving data, while enabling engineers and AI agents to analyze results.
What you'll do
- Build evaluation pipelines and reports to compare model versions, measure performance, and detect regressions.
- Own embedding generation and vector-search pipelines for driving data, and build model-introspection tools.
- Develop tools and workflows using LLMs, VLMs, and agents to analyze driving scenarios and evaluation results.
What they ask for
- Strong Python and proficiency with SQL.
- Design and operation of large-scale batch or data pipelines on cloud infrastructure, including backfills, retries, and cost management.
- Experience building systems to measure or evaluate something, such as metrics, experimentation, evaluation, or analytics platforms.
- Experience with distributed compute and container orchestration, such as Kubernetes or AWS Batch.
Nice to have
- Experience with ML evaluation infrastructure, embeddings and vector search, or model introspection and interpretability tooling.
- Experience using LLMs, VLMs, or agentic workflows for analysis, labeling, or data curation.
Pay
USD $162,000 - $260,000 per year (base pay; Mountain View, CA and Seattle, WA); USD $146,000 - $234,000 per year (base pay; Pittsburgh, PA)
Tools
Python, SQL, Kubernetes, AWS Batch, LLMs, VLMs
AI focus
Apply LLMs, VLMs, and agents to triage, label, cluster, and summarize driving scenarios and evaluation results; build tools for AI agents to analyze datasets and model behavior.
Model-written summary · the original posting is authoritative