Location: McLean, VA
Eligibility: Must be a U.S. Person (required for access to an ITAR / export-controlled
environment)
About the Role
We are hiring a Senior AI / LLM Engineer to design and build LLM-powered features and
applications across a range of use cases. You will work hands-on across the modern AI
engineering stack — retrieval, integration, evaluation, and production hardening — and take
ownership of significant pieces of the system from design through deployment. Retrieval-
augmented generation over large, real-world enterprise data is a prominent part of the work,
alongside platform integration and LLM-driven analysis. You will set technical direction within
your area, make sound trade-offs under ambiguity, and help raise the bar for engineers around
you.
What You’ll Do
• Own the design and delivery of LLM-powered features end-to-end — from problem
framing and architecture through production deployment and iteration.
• Build and tune retrieval-augmented generation (RAG) pipelines over large, heterogeneous
enterprise data — ingestion, chunking, embeddings, indexing, and entity/relationship
modeling — with a focus on retrieval accuracy and closing coverage gaps.
• Design and build data ingestion and indexing pipelines that reliably capture content, map
identities across systems, and support incremental/resumable sync at scale.
• Integrate LLMs (via managed platforms such as Amazon Bedrock) for question answering,
analysis, and other tasks, preserving sessions, sources, and citations.
• Integrate with enterprise platforms and collaboration tools through their APIs, including
SSO/OAuth flows and event-driven bot/app patterns.
• Design permission-bounded access and correct attribution in multi-user contexts, so the
system never surfaces data a user could not already access.
• Establish evaluation practices for retrieval quality and answer correctness, and use them
to drive iteration and catch regressions.
• Add observability, logging, and audit trails, and lead debugging of quality and performance
issues in production.
• Guide and mentor other engineers through design and code reviews, and contribute to
shared standards.
Required Qualifications
• 6–8 years of software engineering experience, with at least 2 years building with LLMs or
applied ML in production.
• Strong proficiency in Python (or comparable) and strong engineering fundamentals —
testing, version control, clean and maintainable code.
• Deep hands-on experience with RAG systems: embeddings, vector databases,
chunking/indexing, and a strong track record diagnosing and improving retrieval quality.• Experience designing and building data ingestion/ETL pipelines over large, messy, real-
world datasets.
• Strong experience integrating third-party APIs into backend services, including
authentication flows (OAuth/SSO) and webhook/event-driven patterns.
• Hands-on experience with LLM APIs (e.g., Anthropic, OpenAI, or similar) and
orchestration frameworks.
• Experience building and relying on evaluations for model and retrieval outputs.
• Experience with knowledge graphs or entity-relationship modeling for retrieval.
• Experience building multi-user or multi-tenant systems with scoped permissions and audit
requirements.
• Familiarity with observability and tracing for LLM or data pipelines.
• A rigorous approach to data access, permissions, and handling sensitive information.
• Experience taking systems to production on a major cloud platform (AWS preferred), and
a track record of owning features independently.
Preferred Qualifications
• Experience with Amazon Bedrock or other managed LLM platforms.
• Experience integrating with enterprise collaboration platforms (chat, wikis, ticketing) via
their APIs.
• MLOps exposure: Docker, CI/CD, production service deployment.
• Bachelor’s or advanced degree in Computer Science, Engineering, or a related field — or
equivalent practical experience.
