SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 180,000 - 220,000 / annual
Comfy is building the AI creation engine for visual professionals, with 4 million users and 60,000+ community-built nodes. This role owns the quality systems that determine whether releases are safe to ship, enabling daily deployment cycles instead of weekly ones.
You'll build infrastructure that proves correctness through machines rather than meetings. Key responsibilities include: designing release gates and automation for daily shipping; solving the combinatorial testing problem across 60,000+ community nodes using property-based and generative testing; building perceptual comparison systems for visual correctness in canvas rendering and node layout; creating evaluation harnesses that distinguish real regressions from acceptable variance in non-deterministic model outputs; establishing continuous verification for third-party node compatibility; leveraging AI agents for exploratory testing while maintaining quality gates; and publishing metrics that track quality improvements.
This is a production engineering role, not a process or QA management role. You'll write infrastructure code that other engineers depend on. The ideal candidate has built testing or verification systems at scale, worked on systems where scripted coverage was impossible and implemented smarter solutions (property-based testing, fuzzing, simulation, generative testing), shipped release gates, and is strong in TypeScript or Python with CI/build system expertise. Experience with graphics/canvas testing, evaluation systems for non-deterministic output, open-source maintenance, or ComfyUI itself is a bonus.
Comfy recently raised $30M at a $500M valuation led by Craft Ventures. The team is small and intense, based in San Francisco, with members from Stability AI and Google. The company serves artists, filmmakers, game studios, and major tech companies including OpenAI, Netflix, Amazon Studios, Ubisoft, EA, and Tencent.
