Founding Design Engineer

About Bretton AI Bretton AI is the leading AI agent platform for financial services. Companies like Robinhood, Mercury, and Gusto trust us to automate mission-critical work, starting with anti-money laundering (AML) and counter-terrorist financing investigations. We've raised over $95M from Greylock, Y Combinator, Thomson Reuters Ventures, and other top-tier investors. We're based in downtown San Francisco and our team comes from world-class organizations like SpaceX, Google, Netflix, Stripe, Plaid, and more. About the Role Every pixel we ship has to earn their trust. That means making sophisticated automation feel legible, making audit trails obvious, and making reviewers faster without ever making them guess what the agent just did. This is Bretton's first Design Engineer hire. You'll own the production layer of the product's interface and serve as the guardian of the design system in code, building the components and patterns the rest of the product is built on, and holding the line on performance, accessibility, and reliability as our product scales. You'll partner closely with our product designer on overall direction, build on insights from research, and work with the engineering team to translate that work into front-end code that's ready to ship. The bar: code that holds up at scale and interfaces that feel considered down to the last interaction. We start by understanding the problem before reaching for a tool. AI and self-serve research are default parts of the workflow, not side experiments. The toolkit is whatever moves the work forward: LLMs for thinking through intent and constraints, Cursor and Claude Code for building, Figma when the design system needs attention. We're looking for someone with the instincts of a designer, the rigor of a front-end engineer, and a working relationship with AI that goes beyond the prompt box. What You'll Do Own the design system in code. Build, maintain, and evolve the components, primitives, and patterns the rest of the product is built on. Make decisions that constrain future work in healthy ways. This is the role's center of gravity. Ship the production version of design-heavy work. Take concepts from prototype to production-grade front-end code. Hold the line on performance, accessibility, and reliability for interfaces that compliance analysts use for hours at a time. Make agent behavior legible. Compliance reviewers need to understand what Bretton's agents investigated, decided, and flagged, often well enough to defend it under audit. Design and build the interface patterns that earn that trust without overwhelming the default experience. Partner with our product designer on direction. Validate ideas through working code, push back when something won't hold up in production, and turn shared decisions into shipped interfaces. Pair with engineering to land work in the codebase. Collaborate with the engineering team on architecture, integration, and the long-term health of the front-end. Your code ships into the same codebase they work in, not a separate one. What we’re looking for Significant production front-end experience, with examples of interfaces you owned from concept through shipped code Fluency in React, TypeScript, semantic HTML, and modern CSS, with strong instincts for component architecture Active use of Claude Code, Cursor, or similar AI-coding tools, with clear views on how to get production-grade output from them A portfolio of working interfaces (live or prototype) that demonstrates visual and interaction craft A systems mindset: interest in building primitives others compose with, and judgment about what belongs in the system versus what stays bespoke Care for performance, accessibility, and reliability as part of how you define craft, not as separate concerns Comfort collaborating with a product designer, engineers, and compliance domain experts who will push back on your work Bonus points Experience contributing to or owning a design system or front-end component library Comfort in Figma for design system work Experience designing for agentic or automation-heavy products where the interface has to expose what the system is doing Background in fintech, compliance, regtech, or other trust-heavy domains Familiarity with PostHog or similar product analytics tools