Frontend Design Engineer – Human-in-the-Loop Interfaces

About the role

We are building agents on open-weight LLMs that operate web applications through the browser. Domain experts supervise these agents via a specialized interface that shows what the agent is doing, why, and allows them to intervene. This role designs and builds that human‑in‑the‑loop interface.

Key responsibilities

• Design and build the screens through which people supervise agents: live view, takeover moments, step‑by‑step review, and demonstration.
• Own the design system and component library in code, including tokens, motion, states, accessibility, and reduced motion, documented in Storybook and used across the product.
• Design every state an interface passes through while an agent works, ensuring each transition clearly shows what changed.
• Build whole features end‑to‑end, server‑side included, and validate them with the users.

Required profile

• Experience authoring and owning a design system in code with tokens, reusable motion and documentation that a product is built on.
• Proven ability to ship interfaces for supervising automated systems, showing what they are doing, why, and how to stop them, for non‑technical expert users.
• T‑shaped skill set: deep expertise in one domain with working breadth in a neighboring area, capable of structuring large, incompletely specified problems and delivering independent results.
• Experience building complete web applications end‑to‑end, including server‑side, with a component library maintained in Storybook or comparable, and daily use of AI coding tools.

Required skills

• TypeScript
• React (or comparable framework)
• Design tokens
• Storybook
• Tailwind
• Radix or React Aria
• Framer Motion, Motion One, GSAP, Rive, Lottie
• WebSockets, WebRTC
• XState, Zustand, Jotai
• Web Audio, speech‑to‑text, text‑to‑speech

What we offer

• Founding‑team scope
• AI‑augmented engineering environment
• Access to on‑premise Nvidia B200 GPUs
• Flexible work environment