About the roleYou'll build the AI systems that power taste evals, tooling, data collection, API and RL environments - from agent architectures and data pipelines to the product surfaces where users interact with our platform. The work skews backend (synthetic data, embeddings, crawling, evaluation systems) but you'll also ship front-end tooling and gamified experiences when needed. Early stage, high ownership, lots of building from scratch.
Types of problems you'd work on- Craft agent harnesses, memory and self-improvement loops
- Design evaluation pipelines and synthetic data generation
- Eval design and grading of unverifiable domains
- Create embedding and retrieval infrastructure that scales to millions of requests
- Build crawling and scraping systems for visual data across the web
- Set up inference serving and APIs for client-facing products
- Develop the tooling and infrastructure that makes everything reliable and fast
- Ship internal tools for data operations and external tools for expert annotators
- Build gamified product experiences: taste quizzes, leaderboards, reward flows
What matters to us- Startup DNA: You've built at early-stage companies (pre-seed to Series C) and operate well in ambiguity
- Real AI building experience: You've shipped agent systems, built with LLMs, and understand the craft - whether through your job, open source, or serious personal projects.
- Genuine curiosity about taste: This problem is hard, nuanced and undefined. You find that energizing, not frustrating.
- Creative problem-solving: We're not optimizing existing systems. We're inventing infrastructure for something that doesn't exist yet.
Bonus points- Open source contributions or personal projects that show you build things because you're curious
- Background at creative companies (Figma, Notion, Canva, Adobe, Runway, etc.) or companies with strong index building/crawling (e.g. Firecrawl, Brave, Luma, Pika) or data (Mercor, Surge, etc.)
