L

Lumbra

Orchestrates multi-agent AI in secure environments

Applied AI Engineer

AI Application EngineeringConfirmed live in the last 24 hours
Full-timeNew York, NY, USA

About the Job

Requirements

  • At least 2 years of professional software engineering experience shipping production code, with strong Python; candidates with 4 or more years may be considered at the senior level.
  • Experience shipping at least one LLM-powered feature to real users and knowing how its effectiveness was measured, including the evaluation, metric, and resulting changes.
  • Hands-on experience with at least one of retrieval pipelines, tool use and function calling, or agent orchestration.
  • Experience designing evaluations for non-deterministic output and iterating on prompts, retrieval, and tooling based on failures.
  • Strong software fundamentals, including testing, code review, typed interfaces, and maintaining code through constant change.
  • Ability to turn an underspecified request into a working feature without a fully specified ticket.
  • Strong fundamentals in scalable distributed systems, service-oriented architectures, networking, and systems design; ideally, experience owning both the technical vision and execution of a foundational platform system end to end.
  • Proficiency in Python, Go, Rust, or a similar systems language.
  • Experience with cloud infrastructure such as GCP, AWS, or Azure, and container orchestration such as Kubernetes.
  • Ability to own work from design through deployment and operations.
  • Ability to navigate ambiguity and make sound technical decisions independently, ideally across both 0-to-1 and more mature team or company settings.
  • Willingness to work on site in New York City.

Responsibilities

  • Build AI features end to end, from prototype to production, including agent behaviors, retrieval, tool use, and the interfaces analysts use.
  • Create datasets, graders, and regression suites that evaluate shipped features and catch capability loss without overfitting to specific outputs.
  • Iterate on prompts, context, retrieval, and tooling based on failure analysis and evidence rather than intuition.
  • Ensure production quality for model calls, including latency, cost, error handling, fallbacks, and observability.
  • Build features that remain effective across model providers, including models available on classified networks.
  • Use feedback from classified deployments, received weekly, to make product fixes.
  • Build and scale the agent loop and execution runtime that sends requests to models, manages tool execution and long-running sessions, and returns results to model context.
  • Build automated benchmarking pipelines and evaluation suites that stress-test accuracy, latency, context-window limits, and regressions.
  • Develop reliable retrieval-augmented generation frameworks and orchestration layers for high-throughput, low-latency AI features.

Desired Qualifications

  • Experience with evaluations, agents, or retrieval pipelines at scale.
  • Experience with Model Context Protocol or similar tool-use frameworks.
  • Experience with open-weight models, self-hosted inference, or constrained compute environments.
  • Experience with graph databases or vector search for knowledge representation and retrieval.
  • Enough TypeScript and React to build the user interface for shipped features.
  • Exposure to intelligence analysis, open-source intelligence, or other analytical workflows.
  • Eligibility to obtain a U.S. security clearance.

About the Company

Lumbra

Orchestrates multi-agent AI in secure environments

Lumbra builds LumbraOS, an operating system for agentic AI in high-consequence environments. It coordinates multi-agent workflows across agencies and systems with full provenance, continuous evaluation, and human oversight on every action. Nebula orchestrates agents and enforces data protections, StarChart provides a customer-owned knowledge graph with sources and sensitivity labels, and Astrolabe tests agents before deployment and scores them in production. The platform is model-neutral and supports classified, air-gapped, and on-premise deployments, and it is deployed on U.S. government networks with plans to extend the architecture to regulated commercial sectors.

Company size
11-50
Company stage
Series A
Total funding
$25M
Headquarters
Arlington County, VA
Founded
2025

Recently Posted Jobs

Explore more opportunities at Lumbra

See more jobs