Staff Software Engineer, Havenly Brands

• You've operated at a staff level and owned architecture and execution for complex systems across more than one team or product area
• Strong full-stack fundamentals across web, services, data, and infrastructure
• You've built or shipped AI-backed product features and can evaluate their quality honestly
• You set technical direction through influence and clear writing, and you stay accountable for systems in production

Desirable

• Architecture for our AI design experiences, from generating and editing room designs to product recommendations, including how we evaluate output quality and keep a human accountable for what ships
• The hardest, least-defined initiatives across the 3D and Studio stack, the AI agent platform, the TypeScript graph layer, and the services behind them
• The path off our older PHP monolith toward the graph and modern services, without stalling product work
• In 90 days: you've gone deep on the platform, shipped real improvements, and taken ownership of one ambiguous, high-impact area
• In a year: you've defined and delivered the architecture for a major AI design capability, raised the quality bar across the team, and mentored engineers into bigger scope
• Alongside our agent platform and LLM tooling, we're building an internal harness so agents can take on real engineering work, like error resolution, bug fixes, and feature enhancements, around the clock
• The team ships with Claude Code, Codex, and Cursor
• We want someone who can help architect this, has built or shipped AI-backed product features, can design eval sets and quality baselines, and can reason about model, data, and latency tradeoffs
• Technologies we use:
• AI: Claude Code, Codex, Cursor, an internal agent platform, and LLM tooling
• Web and mobile: React/TypeScript customer and designer apps, plus a native iOS app in Swift
• Backend: PHP, Node/TypeScript, a TypeScript GraphQL layer, Python
• 3D and Studio: Python rendering pipelines, our in-house 3D toolchain (Lunes), floorplan and visualization systems
• Infrastructure: AWS, Terraform, Dock

• An initial phone screen, followed by two to three rounds of interviews with the team to confirm technical skill, personal fit, and role alignment