SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 165,000 - 200,000 / annual
DEFCON AI is an insights company leveraging AI, mathematical optimization, data analytics, and software engineering to optimize complex systems and help organizations anticipate, assess, and mitigate disruptions.
You will join the analytics and AI engineering team building an AI-assisted platform that ingests records from dozens of external feeds, resolves them to the correct person, surfaces high-priority items for human review, and explains every recommendation in plain, defensible terms. The system runs in a secure government cloud environment and handles problems where precision and detail matter.
As ML Engineer, Retrieval & Grounded Generation, you will own the full lifecycle of retrieval-augmented generation (RAG) systems: building embeddings, vector storage, and retrieval at scale; integrating language models so generated text is bound to cited source records with tested citation guarantees; designing prompts and output schemas; managing model packaging, versioning, serving, and rollback; and instrumenting comprehensive telemetry for retrieval quality, generation quality, grounding failures, latency, and throughput.
Key responsibilities include:
- Implement embeddings, vector storage, and retrieval across a large, provenance-tracked evidence base
- Integrate language models with citation binding; test for citation failures rather than assuming them away
- Design prompts and output schemas
- Own model packaging, versioning, serving, and rollback
- Instrument telemetry for retrieval and generation quality, recommendation/version attribution, overrides, abstentions, grounding failures, latency, throughput, and measurement events
- Provide bounded model assistance for difficult narrative extraction where deterministic processing is insufficient, with every output tied to its source passage
- Supply recorded rule context to every model-assisted step so each output carries exact rule versions and ordered context
- Maintain a modular in-boundary serving path (self-hosted or managed) alongside the primary managed inference service to avoid single-provider dependency
This is a fully remote role with occasional travel to DEFCON AI HQ, customer sites, and vendor facilities as required.
**Requirements:**
- 5+ years of experience, including a production or near-production retrieval-augmented (RAG) system you built yourself
- Ability to speak in detail to your retrieval design: which vector store you used and why, how you tested grounding, what citation failures looked like in practice, and how rollback worked
- Strong Python with hands-on experience (posting text truncated; full requirements not visible in excerpt)
