Senior Software Engineer, Cortex Quality
Recruiter Fit Breakdown & Candid Summary
This is a senior-level, measurement-first engineering role focused on evaluating, optimizing, and scaling Snowflake's AI coding agent, Cortex Code.
The ideal candidate brings 6+ years of production software experience including LLM features, strong Python/TypeScript/Go skills, and a rigorous approach to building evaluation harnesses.
You will work on a high-powered modeling and infrastructure team to turn fuzzy agent behavior signals into hard metrics and reliable production value.
Candidates without hands-on experience shipping AI/LLM features or a systems-thinking approach to evaluation should not apply.
Role Responsibilities
- 1Take AI agents from prototype to production by designing and refining agent behaviors for real coding and data-engineering workflows.
- 2Own agent quality end to end by building eval harnesses, diagnosing failure modes, and running experiments to improve core metrics.
- 3Drive down cost and latency without regressing quality through prompt caching, context compaction, and model routing.
- 4Debug production failures systematically and build regression tests and data pipelines to feed real-world task insights back into evals.
- 5Onboard and evaluate frontier models to measure their strengths and determine their fit within the product.
Skills Matrix
Must-Have Skills
Nice-to-Have Skills
Will an ATS filter reject your resume for this role?
Check your keyword match score, missing must-have skills, and formatting flags — before you apply.
No signups. Free check in 3 seconds.
Ready to submit your application?
Apply directly on Snowflake's official job portal.