Full Stack Software Engineer, Evaluation Tools
$210k - $311k • Sunnyvale, California USA • Full-time
Posted 28d ago
Job Location
Sunnyvale, California USA
Tech Stack
Remote Work Policy
On-site
Employment Type
Full-time
Categories
AI Engineer
About the job
The Evaluation Tools group builds internal products that accelerate the AI Driver development loop, from defining tests to debugging model behavior. Model developers, researchers, and QA engineers rely on these tools to understand driving performance and scale evaluations across millions of scenarios. You will join the Search & Agents squad, developing search, scenario mining, and agentic tooling that allows users to find driving scenarios and create tests without writing SQL. This role involves working across the full stack, owning features end-to-end, from user interaction and idea shaping to building robust software and validating impact. You will have opportunities to define new projects as user needs evolve.
Responsibilities
- Build full-stack features for search, scenario mining, and agentic tooling.
- Collaborate with model developers and researchers to define success metrics before development.
- Scale the scenario mining framework from prototype to production.
- Enhance agent tooling to enable end-to-end evaluation workflows.
- Work with large-scale data systems like Databricks and Spark.
- Ensure the reliability and maintainability of shipped software.
- Contribute to architectural decisions for scalable team products.
- Utilize AI tools to improve development processes and product quality.
Requirements
- Strong Python, TypeScript, and JavaScript development skills.
- Experience with front-end frameworks like React.
- Proficient SQL skills with database design and optimization understanding.
- Experience designing and building production-level full-stack systems and reliable APIs.
- Proven track record of building robust, maintainable systems and applying engineering best practices.
- Excellent communication and collaboration skills, including cross-time zone work.
- Comfortable working independently in a fast-paced, ambiguous environment.
- Experience with search, embeddings, or retrieval systems.
- Experience with large-scale data processing using Spark or similar.
- Experience building agentic or LLM-powered developer tools.