Senior AI and Knowledge Systems Engineer

GeoYeti•Chantilly, VA

2d•$150,000 - $210,000

About The Position

Bcore is hiring a Senior AI and Knowledge Systems Engineer for its Insight Solutions division in Chantilly, Virginia, supporting customer missions. The role focuses on building an enterprise knowledge system that brings together staff reports, plans, and tracked items currently distributed across Jira, Confluence, SharePoint, and other sources. The engineer will develop NLP- and LLM-based workflows that answer natural-language questions in the style used by staff, and will grow an initial working version toward initial and full operational capability. The position involves working closely with end users to validate outputs, conduct user testing, and improve the system based on staff feedback. The engineer will design prompts, APIs, and MCP connections; integrate with real-world systems that may have older versions, limited or unreliable APIs, or no existing connectors; and document workable integration approaches. Access control and information protection are core design requirements: the system must enforce existing permissions, classification markings, compartments, and need-to-know restrictions across its data, indexes, embeddings, prompts, and outputs, while supporting audit logging and keeping data on the appropriate networks. The successful candidate is a hands-on builder who can work through ambiguity, communicate technical decisions to technical and non-technical audiences, and balance AI capabilities with careful validation and data protection. An active TS/SCI clearance with CI poly is required to start.

Requirements

Active TS/SCI clearance with CI poly required to start.
Experience building NLP- or LLM-based applications, including prompt engineering and working with LLM APIs.
Experience designing and building APIs and integrating multiple data sources.
Experience implementing access controls in data or search systems, such as role- or attribute-based access, permission-aware retrieval, and audit logging.
Experience organizing large collections of documents or data so they can be searched and reused.
Ability to explain technical work to non-technical users and leadership.

Nice To Haves

Experience with retrieval-augmented generation (RAG) and vector- or graph-based search.
Experience with Jira, Confluence, or SharePoint, including their APIs and content structures.
Experience with MCP servers or similar LLM tool-integration approaches.
Experience building custom MCP servers or connectors when no off-the-shelf option exists.
Experience integrating with older or on-premises software versions and working around limited, inconsistent, or poorly documented APIs.
Experience with knowledge management, knowledge graphs, or semantic data standards.
Experience designing methods to test LLM output, including accuracy checks and human review.
Experience building and deploying tools on JWICS or higher-classification networks.
Familiarity with intelligence community data handling, including classification markings, compartments, and need-to-know controls.
Familiarity with how military or command staffs produce reports, plans, and operational products.
Strong Python skills and experience with modern data and AI tooling.
Experience working in Agile environments with short, iterative delivery cycles.
Experience running user research or user testing on analytic or data-driven tools.

Responsibilities

Design and build an enterprise knowledge platform that brings together staff reports, plans, and tracked items from Jira, Confluence, SharePoint, and other sources.
Build NLP- and LLM-based workflows that answer natural-language questions with responses written in the style of the staff.
Involve staff in reviewing system outputs, gather feedback, and use it to tune the system and build user trust.
Release selected information to the model for controlled testing to measure accuracy, identify gaps, and understand system behavior.
Observe users performing tasks as they normally would, compare system outputs with their results, and use differences to guide further development.
Build and refine prompts, APIs, and MCP connections that link models to Jira, Confluence, SharePoint, and other data sources.
Enforce user permissions, classification markings, compartments, and need-to-know restrictions throughout the system, including search indexes, embeddings, prompts, and outputs.
Implement audit logging and ensure data remains on the networks where it belongs.
Integrate with real-world systems despite older versions, limited or unreliable APIs, or missing MCP connectors; build connectors or adapt existing integration approaches as needed and document findings.
Grow the first working version into an enterprise knowledge store, working toward initial and full operational capability.