Machine Learning Engineer (Retrieval & Grounded Generation)

DEFCON AI, Inc.; United States; On-site; USD 150,000 - 190,000; Full time

Benefits offered by this job

Top-of-market salary, bonus, equity

Employer-paid health insurance

Unlimited PTO

Parental leave

Travel to DC or other offices

Professional development

Job summary

DEFCON AI, Inc. is seeking an ML Engineer focused on Retrieval & Grounded Generation in the United States. You will build embeddings, vector storage, and retrieval at scale, and integrate language models so that every output is bound to cited sources with tested citation failures.

You will own prompts, output schemas, model packaging, versioning, serving, and rollback, plus telemetry to measure retrieval and generation quality. Active US Secret clearance and citizenship required.

Qualifications

• Strong Python with hands-on experience in embeddings and vector retrieval at scale.
• 5+ years of experience in production or near-production retrieval-augmented (RAG) systems built by you.
• Ability to discuss retrieval design, vector store choices, grounding tests, and rollback implementations.
• Active US Secret clearance required to start; US Citizenship required.
• Experience deploying models into restricted or air-gapped environments.
• Self-hosted or open-weight model operation.
• Fine-tuning, adapters, or custom embeddings.
• Federal DevSecOps, RMF, ATO, or DoW cloud environment experience.

Responsibilities

• Build embeddings, vector storage, and retrieval at scale with provenance tracking.
• Integrate language models to bind generated text to cited sources and test for citation failures.
• Design prompts and output schemas; own packaging, versioning, serving, and rollback.
• Instrument telemetry for retrieval and generation quality and measurement events.
• Provide bounded model assistance with outputs tied to source passages.
• Ensure the platform remains self-hosted or managed across providers.

Skills

Python

Embeddings

Vector retrieval

RAG system

Model deployment

Grounded generation

Retrieval design

Security clearance

Air-gapped environments

Tools

Vector store

Open-weight models

Job description

• As ML Engineer, Retrieval & Grounded Generation, you’ll build embeddings, vector storage, and retrieval at scale, and integrate language models so that every piece of generated text is bound to cited source records and citation failures are tested for rather than assumed away
• You’ll also own prompt and output-schema design; model packaging, versioning, serving, and rollback; and the telemetry hooks that make later measurement possible without manual reconstruction - real infrastructure for a real production system, not a demo
• Implement embeddings, vector storage, and retrieval across a large, provenance-tracked evidence base
• Integrate language models so generated text is bound to cited source records; 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 defined with Model Test
• Provide bounded model assistance for difficult narrative extraction where deterministic processing is insufficient, with every output tied to its source passage
• Supply the recorded rule context to every model-assisted step, so each output carries the exact rule versions and ordered context it received
• Maintain a modular in-boundary serving path, self-hosted or managed, alongside the primary managed inference service, so the platform does not depend on one provider’s availability or approval
• What Success Looks Like:
• Generated explanations that assert no more than the sources support, with the citation path intact and citation failures tested rather than assumed away
• A retrieval system that performs at scale on a large, provenance-tracked evidence base
• Model rollback that works when it’s needed, with telemetry complete enough that measurement does not require manual reconstruction

Benefits

• Top-of-market competitive salary, bonus, and equity package
• 100% employer paid comprehensive health insurance including medical, dental and vision for you and your family
• Unlimited PTO, with your manager’s approval
• 14 weeks of fully paid parental leave
• Occasional travel to our offices in Washington D.C or alternative locations to collaborate with your team in person
• Opportunities for professional development and continued learning

Strong Python, with hands-on experience in embeddings and vector retrieval at scaleClarity on what actually shipped in past work — prototype, proposal, or deployed code — since that distinction matters more here than the title on a resume5+ years of experience, including a production or near-production retrieval-augmented (RAG) system you built yourselfAbility 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 workedActive US Secret clearance required to startUS Citizenship RequiredExperience deploying models into restricted or air-gapped environmentsSelf-hosted or open-weight model operationFine-tuning, adapters, or custom embeddingsActive Top Secret clearanceFederal DevSecOps, RMF, ATO, or DoW cloud environment experience