Problem Statement: Legacy software infrastructure across the Department of the Air Force (DAF)—particularly within satellite ground control and command and control (C2) systems—imposes compounding technical debt. Outdated Python 2 and Java codebases, unsupported multi-generational dependencies, and the absolute mandate for fully disconnected operational environments make traditional brownfield migration prohibitively slow. This technical debt severely delays secure cloud migration, strains sustainment budgets, and introduces unacceptable risks to mission readiness.

The Blueprint: This session introduces the Mission Agent Mesh, a distributed, multi-agent AI architecture engineered to accelerate legacy system transformation. Drawing from successful defense implementations, we outline a coordinated orchestration framework of specialized agents powered by Small Language Models (SLMs) running locally in secure, air-gapped enclaves. Utilizing emerging Model Context Protocol (MCP) and Application-to-Application (A2A) standards, these discovery-ready agents securely collaborate to analyze legacy source code, map dependencies, identify deprecated APIs, generate container-ready refactored equivalents, and automatically produce characterization tests to guarantee strict behavioral equivalence.

Operational Impact: Attendees will examine a scalable blueprint demonstrating how deterministic agentic orchestration compresses modernization timelines from years to months, allowing organizations to out-pace adversary development cycles. We will share measurable outcomes and practical lessons learned from defense deployments directly applicable to Air Force and Space Force operational architectures, enabling a repeatable path toward continuous software refactoring at the speed of relevance.