Baltimore, MD
Powered by AI/ML
DEBRIEF will enable Redshred to make manual, time-intensive quarterly reporting cycles more efficient with AI-powered data extraction from documents and automated briefing material generation across the Defense Logistics Agency's R&D portfolio.
FOR IMMEDIATE RELEASE
[BALTIMORE, MD] — [27 APRIL 2026] Redshred, a leader in artificial intelligence solutions for complex document understanding, has been awarded a Phase II Small Business Innovation Research (SBIR) contract by the Defense Logistics Agency (DLA) to develop DEBRIEF (Data Extraction for Briefing Review and Intelligent Enterprise Feedback), a production-grade document intelligence system designed to transform how the U.S. Department of Defense conducts Program Management Reviews.
Program managers across DLA’s R&D portfolio spend hours each quarter manually compiling status reports and assembling briefing materials from disparate sources, such as PowerPoint quad charts, Word status reports, and PDF deliverables, before populating the Strategic Portfolio Management (SPM) system. DEBRIEF eliminates this burden.
"The challenge DLA faces isn't a lack of data. It's that the data is locked inside documents that no system can read automatically,” said Jim Kukla, CTO of Redshred. “DEBRIEF changes that. We built RQL and our dynamic ontology layer specifically so that organizations can get structured, traceable answers from their existing documents without writing a single line of custom integration code."
Built on Redshred’s platform, DEBRIEF introduces three core innovations: dynamic ontology mapping that adapts to diverse document formats without custom coding; unstructured document ingestion capable of processing PowerPoint, Word, and PDF source files at scale; and deterministic document querying via the proprietary Redshred Query Language (RQL), which enables traceable, auditable data extraction that meets the stringent reliability standards required for defense applications.
Phase II development will have two prototype delivery cycles. The first establishes requirements discovery and a Minimum Viable PMR (MV-PMR) template, followed by deployment of the core extraction pipeline into a DLA cloud environment with three-tier dynamic ontology mapping and full SPM integration. The second cycle delivers an interactive portfolio planning dashboard, automated PowerPoint briefing generation, and comprehensive change management and training for DLA R&D personnel — culminating in an end-to-end system demonstration validating all technical objectives.