sr. front-end engineer

About Osmo

Osmo is building the AI-native motion graphics studio.

We help frontier startups, researchers, and creative teams turn ideas, designs, and raw assets into polished launch films, product demos, and cinematic explainers. Faster. With more taste. With real creative control.

Our tools already support launches for companies like Parallel, Basis.ai, and YC-backed teams, and that real-world use shapes everything we build. We’re backed by top venture capital firms including Bain Capital Ventures, MVP Fund, and Collaborative Fund.

The role

The Osmo editor is where ideas become motion. It’s a real-time, AI-native canvas: a chat agent generating React/Remotion code, a timeline with keyframeable properties, an inspector for every item type, and a preview that updates as code streams in. It’s the surface our users live in, and the bar for craft is high.

We’re looking for a Senior Front-End Engineer to own and push that surface forward. You’ll build new tools, design custom UI for problems off-the-shelf components can’t solve, and tighten the loop between an AI agent writing code and a human shaping a film.

We want one of the best front-end engineers in the world. Someone who’s spent years thinking about pixels, frames, and how it feels to drag a clip across a timeline.

What you’ll do

• Develop new features in the Osmo editor (timeline, canvas, inspector, agent chat) that are sophisticated, maintainable, and scale to projects with hundreds of clips, keyframes, and assets.
• Build custom UI components where Radix and shadcn don’t go far enough: timeline interactions, keyframe editors, drag/drop systems, code streaming surfaces, Figma import flows.
• Go low-level when the problem demands it: Canvas, WebGL, GPU-accelerated rendering, custom hit testing, frame-accurate scrubbing, transform math.
• Design and ship the human side of the dual-agent system: how users invoke agents, watch code generate live, intervene, and iterate.
• Optimize playback, scrubbing, and re-render performance so the editor stays buttery even with heavy compositions.
• Extend the item system (Animation, Text, Image, Video, Captions, Audio, etc.), adding new item types end-to-end across types, layer renderers, inspector panels, timeline previews, and AI tool schemas.
• Work closely with the AI/agent engineers to evolve tool schemas, streaming protocols, and the CONTROLS pattern that lets users tweak generated animations.
• Partner with designers and motion artists to translate Figma into 1:1 implementations and grow our design system as the product surface expands.

What you’ll bring

• Background in motion design tools or pro creative software: After Effects, Premiere, DaVinci Resolve, Final Cut, Figma, Rive, Cavalry, Houdini, or similar. You’ve either built one, lived inside one, or both.
• A track record of being one of the best front-end engineers in the room. You’ve shipped polished, complex products and the bar you set for craft is uncomfortable for everyone else.
• Low-level chops. Comfortable with Canvas, WebGL/WebGPU, requestAnimationFrame loops, custom rendering pipelines, transform matrices, and the kind of performance work that doesn’t fit in a React component.
• Design system experience at scale. You’ve built and maintained component systems used by many engineers across many surfaces, and know when to extend tokens vs. when to break out.
• Deep React + TypeScript. Hooks, concurrent rendering, performance tuning, correct types over workarounds. React 19 / RSC familiarity is a plus.
• Custom UI from scratch (linear timelines, virtualized canvases, drag interactions, keyframe curves, gesture systems) when no library fits.
• Strong product instincts. You think about the user’s task and the feel of the tool, not just the ticket.
• Excellent communication. Small team, lots of cross-functional surface area.

Nice to have

• Experience with Remotion, video editors, NLEs, animation tooling, design tools, or DAWs.
• Shipped Vercel AI SDK, tool-calling, or agent UIs.
• Familiarity with our stack: TanStack Start / Router, Drizzle, Tailwind v4, Radix, Biome, Better Auth, Neon.
• A portfolio of motion, generative, or interactive work: code, films, or otherwise.

Location

San Francisco preferred. We work in person regularly and there’s no replacement for sitting next to designers and engineers while building a creative tool. Open to exceptional remote candidates willing to travel for on-sites.

Compensation

Competitive salary base + meaningful equity.

We offer comprehensive health, dental, and vision coverage, a flexible work environment, and equity that reflects the early-stage upside of building a category-defining tool. Compensation may vary based on experience, skills, and location.