What happens when thousands of federal officers can't find the policy, the precedent, or the intelligence they need at the moment they need it? Decisions slow down, and at a border security agency, slow decisions have real consequences. IPNS is fixing this. We're deploying an AI-powered enterprise search and retrieval platform at a large federal law enforcement agency, transforming how thousands of personnel find and use mission-critical information scattered across siloed systems. We're looking for a hands-on engineer to lead the technical implementation of our agentic RAG platform, working directly with our CEO and the customer's technical team. This is a ground-floor role on a high-visibility federal AI program: you'll own the search and retrieval stack end to end, from data ingestion through relevance tuning, on one of the most consequential enterprise AI deployments in the federal space.

Core responsibilities:

• Lead hands-on implementation of the enterprise search platform in the customer's cloud environment, working alongside IPNS leadership and customer engineering teams
• Build and tune data ingestion pipelines across diverse sources (SharePoint, network drives, structured databases), including document parsing, chunking strategy, metadata extraction, and sensitivity classification
• Implement and optimize the embedding and indexing layer: Vertex AI embedding models, Elasticsearch vector search (dense_vector/kNN), and federated multi-index retrieval
• Own retrieval quality: relevance tuning, hybrid search (semantic plus keyword), evaluation frameworks, and iterative improvement based on user feedback
• Configure and extend the multi-agent orchestration layer (query decomposition, specialized domain agents, response validation) and integrate with the customer's existing chat interface
• Ensure the implementation meets federal security and compliance requirements (FedRAMP High alignment, role-based and document-level access controls)
• Support POC milestones, customer demos, and technical working sessions with agency stakeholders and prime contractor partners

Key qualifications/experience:

• 5+ years in search/information retrieval, ML engineering, or data engineering, with 2+ years building LLM/RAG systems in production (not just prototypes)
• Significant Elasticsearch (or similar) experience: index design, relevance tuning, vector search; cross-cluster search a strong plus
• Google Cloud AI stack: Vertex AI (Vector Search, embedding models, Gemini), plus supporting services like Dataflow, Document AI, and GKE
• Strong Python and data manipulation skills: ETL pipelines, regex, parsing messy structured and unstructured data at scale
• Practical understanding of embeddings, chunking strategies, hybrid retrieval, and RAG evaluation methods
• Experience with agent frameworks or multi-agent orchestration a plus
• US citizenship required; must be able to obtain and maintain a federal background investigation. Active clearance or prior DHS suitability determination a strong plus
• Prior federal or regulated-environment delivery experience a plus
• Self-directed and comfortable in a small-company environment where you'll wear multiple hats

Why IP Network Solutions (IPNS)

At IPNS, you will work on high-impact national security and border protection missions, delivering cutting-edge AI/ML solutions that directly support CBP operators and decision-makers. This role offers the opportunity to shape AI capabilities from inception to operational deployment in a real-world federal mission environment.

Job Type: Part-time

Pay: From $175,000.00 per year

Application Question(s):

• Do you have experience working with MLOps, containerization (Docker), and orchestration (Kubernetes)?

Security clearance:

• Top Secret (Required)

Work Location: Hybrid remote in Ashburn, VA