Staff Product Software Engineer

LumaRedwood City, CA

About The Position

You'll set technical direction for how products get built at Luma, directing AI coding agents to ship production systems fast while making the disciplined calls on scope, structure, and trade-offs. This is a senior role where judgment, not typing, is the constraint. With AI making implementation cheap, weak decisions compound faster, so the job is deciding what to build, what not to, and where code should exist at all. It's for a technical leader with deep system-design judgment and real product intuition who is genuinely world-class at directing AI agents. If you measure your impact by lines of code written, this isn't framed for you.

Requirements

  • 8+ years building and shipping production software.
  • Multiple non-trivial projects where you owned technical direction and execution end to end.
  • High-stakes technical decisions that materially affected a system or product.
  • Fluency with AI coding tools, used to ship real functionality in production or substantial side projects.
  • Strong technical depth (system design, distributed systems, security, infrastructure) or strong product intuition (user empathy, prioritization, UX taste), ideally both.

Nice To Haves

  • Personal responsibility for a measurable real-world success — a product at 100k+ users or 1M+ monthly actives, a system serving millions of requests a day, or an open-source project with 10k+ stars or meaningful adoption.

Responsibilities

  • Direct AI coding agents to build production systems quickly, holding the line on system design, ownership, and real-world behavior.
  • Decide what to build and what not to build as implementation costs drop.
  • Set technical direction across engineering: system boundaries, data models, ownership, and operational expectations.
  • Co-own product direction with design and research.
  • Make and stand behind hard trade-offs between speed, reliability, complexity, and extensibility.
  • Shape architecture early, and own critical outcomes end to end from framing to deployment.
  • Mentor engineers on using AI tools without lowering rigor, and eliminate recurring classes of design mistakes.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service