Full-Stack Software Engineer, Core Platform
- Compensation
- Not disclosed
- Location
- Arlington
- Work setup
- Onsite
- Seniority
- Not specified
- Employment
- Contract
- Experience
- Not specified
- Category
- Fullstack
- Posted
Skills
Must have
- Ships production Python (async, relational modeling, layered services) and production React and TypeScript
- Reasons comfortably about data flow through every layer, from schema to UI state
- Writes clean, well-tested code and works test-first
Nice to have
- Agentic AI systems: harness engineering, context engineering, or building on LLM APIs
- Distributed task orchestration systems
- Graph databases or vector search for knowledge representation and retrieval
- Model Context Protocol or similar tool-use frameworks
- Identity management (OAuth2/OIDC)
- Open-source developer-tooling contributions
- Intelligence analysis, OSINT, or analytical workflows that inform product intuition
- Cloud infrastructure (GCP or equivalent): managed Kubernetes, managed databases, secrets management What We Offer
- Health, dental, and vision insurance. 99% of your premiums covered, and 100% of your dependents' premiums.
- Optional HSA plan. Lower premium, HSA eligible, fully portable, self directed or managed.
- Health Reimbursement Account for qualified healthcare expenses.
- Life insurance of $150,000 plus AD&D, 100% company paid.
- 401(k) with a 100% match on the first 6%, immediately vested. No cliff, no waiting period. Yours from day one.
- Wellhub Gold. Everything from Gold's Gym to Lifetime, fully covered, plus [solidcore], F45, and hundreds of other studios.
- Unlimited PTO. Real time off, not accrual based. Take what you need, no counting required.
- Every federal holiday, plus additional paid days off over the winter break.
- Supportive leave of absence program, including time off for military service, medical events, and parental leave.
- Breakfast, lunch, and endless snacks every day you are in the office.
Full description
About Lumbra
Lumbra builds the vendor-neutral operating system for agentic AI in high-consequence environments. In our first ten months we won four U.S. Government prime contracts in open competition, every one deploying into classified spaces, and we are live on JWICS. We were selected from a field of more than 200 companies and won, beat two established defense primes head to head, and earned the follow-on now in execution. Lumbra is a team of engineers and operators from the intelligence community, special operations, and frontier AI research, with degrees from MIT, Carnegie Mellon, and West Point, and careers built at Palantir, Anduril, Northrop Grumman, Onebrief, JSOC, NSA, and CIA. They have lived this problem from the inside.
About the Role
Nebula turns days of manual synthesis into minutes of analysis. We want a full-stack engineer to own features end to end across the Python backend and the React frontend, putting real AI tooling directly in analysts' hands.
What You'll Own
- Design and build FastAPI endpoints backed by PostgreSQL and SQLAlchemy for agent configuration, workflow orchestration, and evaluation pipelines.
- Implement interactive React and TypeScript features: conversation interfaces, workflow visualization, and knowledge-graph exploration.
- Build durable workflow orchestration for long-running agent execution, including delegation, monitoring, and timeout propagation.
- Own features across the full stack, from database migration through API contract to UI state.
- Bring clean abstractions and well-defined interfaces to a fast-moving codebase, writing code that survives constant change and stays easy to understand, refactor, and extend.
- Write contract-driven tests across backend and frontend that cover the spec rather than the implementation, working test-first.
- Design testing for non-deterministic AI components: metamorphic testing, property-based testing, and regression suites that catch capability loss without overfitting to specific outputs.
- Contribute to the Python and TypeScript client SDKs so programmatic consumers get ergonomic, well-typed access.
- Own performance across the stack and profile before guessing: API latency, query efficiency, render performance, and end-to-end streaming latency. Treat slowness as a defect.
- Debug across service boundaries with local environments and distributed tracing.
Required Qualifications
- Ships production Python (async, relational modeling, layered services) and production React and TypeScript
- Reasons comfortably about data flow through every layer, from schema to UI state