SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 112,000 - 149,000 / annual
Anduril Industries is seeking an Applied LLM Systems Engineer to design, build, and operate production AI systems that improve how technical documentation is created, transformed, validated, maintained, and delivered across defense and aerospace programs.
You will own the AI-assisted authoring, automated publishing, and multi-agent coordination tooling for the Technical Publications pipeline. This is a systems-engineering role for someone with proven experience delivering production AI systems in enterprise environments. You will build and maintain LLM applications that handle structured content, strict review requirements, security boundaries, and real operational consequences.
Key responsibilities include:
- Design and build production-grade LLM applications for documentation and knowledge-work workflows
- Design orchestration for multi-step workflows that safely coordinate models, tools, and deterministic services
- Optimize context management, token usage, caching, and workflow design for cost, latency, and task success
- Build retrieval, structured-output, and tool-integrated systems that safely interact with enterprise content, source control, issue tracking, and documentation systems
- Design evaluation frameworks and regression testing for prompts, models, tools, retrieval pipelines, and end-to-end workflows
- Establish observability, auditability, rollback, and safe failure modes for AI-assisted workflows
- Work directly with technical writers, illustrators, engineers, and documentation leadership to identify high-value AI applications
- Preserve provenance between source material, generated output, validation results, and human decisions
Required: 5+ years professional software engineering with production services; direct experience delivering at least one production AI system with measurable user adoption; experience optimizing cost, latency, and context usage in production LLM systems; experience designing evaluation, observability, rollback, and failure-handling for nondeterministic systems; strong Python proficiency and modern API/data/service-development practices; sound judgment about where probabilistic AI is appropriate versus deterministic software or human review; eligibility for Secret and higher US security clearance.
Preferred: multi-step orchestration experience; RAG, structured outputs, tool calling, or production LLM workflow patterns; AI integration into enterprise applications; defense, aerospace, manufacturing, robotics, autonomy, or regulated environment experience; technical documentation, structured authoring, or content management familiarity; S1000D, DITA, MIL-STD-40051 knowledge; containerized services, cloud infrastructure, CI/CD, or production monitoring experience.
