AI Engineer, RL & Evals (US)

We are looking for an AI Engineer, RL & Evals with 3-10 years of experience to build reinforcement learning environments, design evaluations, and ship production infrastructure for subjective, creative domains. You'll sit at the intersection of backend engineering and applied ML, this isn't a pure research role, but you'll need deep familiarity with RL and evals to build the systems that make AI good at design. Taste Labs works directly with top frontier AI labs and leading application-layer companies, and you'll have massive ownership over the environments and evaluation frameworks that power both sides of that work. This is a rare opportunity to tackle one of AI's hardest unsolved problems at a fast-moving, well-funded seed-stage startup.

What will you be doing?

• Build and scale reinforcement learning environments and evaluation frameworks for creative, non-verifiable domains like design

• Design unique tasks and grading rubrics that capture elements of "taste" and design quality without humans in the loop

• Build agent harnesses and context layers that power Taste Labs' Brand API and broader product infrastructure

• Collaborate with internal research teams and top frontier labs on crafting environments to improve frontier model output quality

• Ship production code end-to-end, this is a product-oriented engineering role, not research in isolation