Frontend Agent Engineer

Frontend Agent Engineer About the Role

We are building an AI application community where anyone can create, and everyone can benefit. The Frontend Agent is the user-facing layer of this vision-it transforms a single natural language prompt into a complete application creation experience that users can understand, edit, preview, and publish.

This role is not about building traditional web pages. Instead, you'll design and build the interaction system that enables humans and AI Agents to create software together.

You will own the complete frontend experience across the entire Agent workflow:

Prompt Task Planning Tool Call Code Editing Diff Preview Sandbox Rendering Error Recovery Deployment

Your responsibility is to make the Agent's execution process visible, controllable, and trustworthy-allowing non-technical users to understand what the Agent is doing while giving technical users the ability to inspect, intervene, and debug every step.

This is not simply a UI implementation role-you will help define the core interaction paradigm for the next generation of AI-native application builders.

What You'll Build Agent Creation Experience

Build our core App Builder experience, including:

• Conversational input
• Task planning visualization
• Tool execution flow
• File explorer
• Code diff viewer
• Live preview
• Version rollback
• Publishing workflow

Tool Call Visualization

Transform Agent tool executions-including Read, Write, Edit, Search, Build, and Deploy-into intuitive visual states, progress indicators, and user-friendly feedback.

Real-Time AI Generation Experience

Design responsive interfaces capable of handling:

• LLM streaming
• Build logs
• Sandbox status
• Runtime errors
• High-frequency event streams

Ensure the UI remains smooth, resilient, and recoverable throughout long-running AI tasks.

Code & Preview Collaboration

Design seamless interactions between:

• Code editing
• Component-level modifications
• Partial regeneration
• Preview refresh
• Error localization

Build a closed feedback loop between code generation and visual application rendering.

Frontend State Architecture

Design and manage complex application state across:

• Tasks
• Files
• Messages
• Tool calls
• Sandbox sessions
• Deployments
• Version history

Maintain consistency throughout long-running, multi-turn AI generation sessions.

Agent Failure Experience

Present build failures, runtime errors, and tool failures in ways that are both understandable and actionable.

Drive automatic recovery whenever possible while providing meaningful user guidance when intervention is required.

Generated Output Validation

Help build frontend-side quality validation systems covering:

• Visual completeness
• Responsive layouts
• Interaction usability
• Preview and deployment consistency

What We're Looking For Required Qualifications Frontend Engineering

• Strong proficiency in React and TypeScript.
• Experience architecting and shipping complex frontend applications from 0 to 1.

Modern Frontend Stack

Hands-on experience with:

• Next.js
• Vite
• Tailwind CSS
• Zustand
• TanStack Query
• pnpm
• CI/CD pipelines

Frontend Architecture

Deep understanding of:

• Complex state management
• Streaming UI
• WebSocket
• Server-Sent Events (SSE)
• Event-driven architectures
• Long-running task synchronization

LLM Agent Fundamentals

Understanding of modern AI Agent systems, including:

• Tool calling
• Agent loops
• Context management
• Multi-step task execution
• Self-repair mechanisms

Developer Products

Experience building coding products or developer tools.

Familiarity with interfaces such as:

• File explorers
• Code editors
• Diff viewers
• Terminal logs
• Build errors
• Live previews

Product & UX Thinking

Ability to transform complex system states into experiences that feel:

• Clear
• Trustworthy
• Easy to understand
• Low cognitive load

Strong product judgment to recognize why an AI generation experience feels unreliable, confusing, or frustrating.

AI-Native Development

Regular user of modern AI coding tools.

Hands-on experience building production projects with tools such as:

• Claude Code
• Cursor
• OpenAI Codex
• v0
• Bolt

Cross-Stack Ownership

Comfortable working across boundaries between:

• Frontend
• Agent Runtime
• Sandbox Infrastructure
• Backend APIs

Ability to proactively drive end-to-end product execution.

Preferred Qualifications

Candidates with multiple of the following experiences will stand out:

• Experience building AI coding products, website generators, low-code/no-code platforms, online IDEs, or developer tools.
• Built VS Code extensions, Chrome extensions, CLI tools, code editors, Agent debugging panels, or workflow builders.
• Familiarity with:
  • Monaco Editor
  • CodeMirror
  • AST manipulation
  • LSP diagnostics
  • Codemods
  • Symbols
  • References
• Experience building:
  • Tool call visualizations
  • Trace viewers
  • Agent execution timelines
  • Debug consoles
• Experience with:
  • MCP
  • Agent Skills
  • Agent-to-Agent (A2A) communication
  • Multi-Agent collaboration
  • Agent marketplaces
  • Agent plugin ecosystems
• Experience building:
  • Sandbox preview environments
  • WebContainer
  • E2B
  • Vercel deployment pipelines
  • Online build systems
• Maintains production-grade open-source projects, browser extensions, CLI tools, or independent products with active users.
• Experience with generative UI and creative coding technologies such as:
  • p5.js
  • Canvas
  • SVG
  • Framer
  • Three.js