We build consumer AI products downloaded by millions. We ship fast, work close to the newest models, and build as a team: engineers, designers, and product working side by side from the first prompt to production.

We're hiring a Software Engineer whose core focus is turning LLM capabilities into product: agentic pipelines, tool calling, streaming — the layer where AI actually becomes a feature people use daily.

What You'll Do

• Build and ship AI/LLM features end-to-end: API integration, prompt & system design, tool/function calling, multi-step agent pipelines, streaming responses.
• Make those features production-grade: fast, reliable, safe, and cost-efficient at scale.
• Design and maintain backend services in Node.js + TypeScript, backed by Firebase (Firestore, Auth, Cloud Functions).
• Hunt down bottlenecks across the backend and inference layers — latency, memory, and model/infra cost.
• Contribute to our Flutter apps when product work spans client and backend.
• Work directly with product and design to take features from concept to launch.

What We're Looking For

• Hands-on LLM experience is the core requirement — prompt engineering, tool calling, and agentic workflows in real products, beyond a single chat call.
• An AI-native way of working: you prototype fast with AI dev tools and reason about model behavior, strengths, and limits.
• 2+ years of professional software development experience.
• Strong backend engineering with Node.js and TypeScript (services, async, streaming).
• Experience with Firebase, and a solid grasp of architecture, scalability, and performance.
• Working proficiency in Flutter/Dart — or the appetite to ramp up quickly.

Nice to Have

• GCP/AWS, containers, serverless
• Sandboxed code execution
• Native mobile (Swift or Kotlin)