Forward Deployed Engineers turn what Luma's models can do into systems customers actually rely on.
You embed with a customer, learn their workflow, define the problem with them, and build the production system that solves it, with their data and their constraints.
You own both the relationship and the code.
This is engineering at the edge, where ambiguity is the default and "what should we build?" is yours to answer.
It's built for people who've already been the single point of ownership on a hard problem: a technical founder, the engineer who flew out and rebuilt a customer's workflow, a consultant who actually wrote the system.
If you need a PRD and a well-defined ticket queue, this isn't it.
What You'll OwnWork directly with customers to understand their workflows, constraints, and goals, and define the problem yourself.
Build production systems on Luma's APIs, models, and internal tools.
Prototype, ship, iterate, and maintain systems customers depend on in real environments with real data.
Build around platform gaps when something's missing, and feed recurring patterns back to platform and R&D to productize.
Use AI to accelerate implementation, while the core of the work stays engineering judgment: what to build and how to make it reliable.
First 90 DaysOne way the first 90 could unfold.
Days 1–30 — Immerse & Diagnose: Embed with a first customer, learn their workflow, and define the real problem.
Days 30–60 — Ship & Validate: Build and deploy a production system they actually use.
Days 60–90 — Scale & Systemize: Spot the pattern worth productizing and work with R&D to turn it into platform.
What You Bring2+ years of professional software engineering, with software shipped end to end.
Strong proficiency in at least one general-purpose language (Python, TypeScript, Go, or similar).
Ability to work across the stack and learn unfamiliar systems quickly.
Bachelor's in Computer Science or equivalent practical e.