Baltimore, MD
Powered by AI/ML
Baltimore, MD
Powered by AI/ML
Company News

Redshred Secures SIFTED Phase II SBIR to Scale AI-Powered Maintenance Intelligence for the Navy

April 13, 2026

FOR IMMEDIATE RELEASE

BALTIMORE, MD — APRIL 13, 2026. Redshred, a leader in artificial intelligence solutions for complex document understanding, has been awarded a Navy Phase II SBIR to continue development of its Semantic Information for Faults, Troubleshooting, and Extraction from Documents (SIFTED) system, advancing from proof-of-concept to prototype. 

Phase I successfully demonstrated the feasibility of AI-driven document analysis for local engineering instruction generation, and Phase II will expand the system into a production-grade platform that can operate on Navy RDT&E networks and integrate with Navy maintenance systems, creating time and cost savings at Fleet Readiness Centers (FRCs).

Adding to Redshred’s COACH solution for rapid instruction authoring and AI-guided instruction execution, SIFTED automates some of the most labor-intensive steps in the maintenance data lifecycle: localizing relevant pages across complex repair documents, assembling technical data packages (TDPs), and generating detailed repair instructions.

Generating repair instructions at FRCs currently takes weeks to months. SIFTED can reduce that timeline down to minutes or hours, with a projected annual labor savings of more than $5.5M at FRCSW alone. The system's offline capability also supports Aircraft Battle Damage Repair (ABDR) use cases where connectivity cannot be guaranteed.

"SIFTED Phase II represents a major milestone for Redshred and for the future of naval aviation maintenance," said Jim Kukla, CTO of Redshred. "We're not just accelerating a process. We're rethinking how maintenance knowledge is captured, structured, and put to work. SIFTED turns static documents and historical repair data into an active, intelligent system that technicians and engineers can act on immediately."

Phase II development will focus on AI/ML-based data extraction and automated repair instruction generation and traceability. The program is designed to demonstrate SIFTED's full feasibility at scale and lay the groundwork for transition across the broader defense enterprise, including NAVSEA, USAF Air Logistics Centers, US Army Aviation, US Coast Guard, DLA, DHS, and NASA.

Redshred will be looking to expand these capabilities into commercial markets, including manufacturing, aerospace MRO, automotive, and industrial maintenance.

‍

For more information about Redshred, please visit Redshred.com or contact:

Jeehye Yun
hello@redshred.com

‍

About Redshred:

Redshred is a leading provider of AI-powered document intelligence solutions for federal agencies and commercial enterprises. Redshred’s patented technologies redefine conventional data management by turning technical documents into a database by combining layout geography recognition, content segmentation, and strategic deployment of best-in-class COTS, open source, or custom ML models—avoiding single-model limitations. Their state-of-the-art artificial intelligence revolutionizes document processing and analysis.