Nvidia Senior Software Engineer Evaluation Metrics - Autonomous Vehicles 

United States, California 

923227154

• Manage and implement a product critical metric and evaluation workstream within the broader AV validation strategy

• Develop robust methods for analyzing sophisticated, unstructured testing with real world driving

• Deliver metric systems, dashboards, and analysis tools that drive product decisions

• Employ data science methods like debiasing, sampling, and exposure modeling to guarantee accurate, significant critical metrics

• Collaborate closely with simulation, infrastructure, operations, and engineering teams to embed your metrics in daily development workflows

• Integrate on road performance with upstream simulation efforts, ensuring follow through between real-world outcomes and test coverage

• Influence metric adoption and alignment across teams by providing clear, consistent technical mentorship

What we need to see:

• 5+ years of experience in Data Engineering and Data Science Roles

• BS/MS in Computer Science or equivalent experience

• Demonstrated experience in developing and delivering critical metrics systems or data products for complex, real-world or ML-based systems

• Hands-on application of statistical techniques such as debiasing, sampling, exposure control, or experimental design

• Proficiency in Python, SQL, and scalable data tools (e.g., Spark, BigQuery)

• Strong product approach with a focus on usability and measurable impact

• Tight-knit collaboration and communication across technical teams

• Proficiency in articulating evaluation reasoning and results clearly to both technical and non-technical customers

Ways to stand out from the crowd:

• Background in autonomous vehicles, robotics, or simulation domains

• Experience with simulation feedback loops tied to real-world performance

• Familiarity with performance validation for safety-critical systems in unstructured validation environments

• Exposure to labeling pipelines or automated evaluation tooling. Experience with both human-labeled and automated critical metrics

• Taking ownership of a data science workstream from strategy through implementation, ideally involving multi-functional coordination