The Center of Excellence in Wireless and Information Technology (CEWIT) at Stony Brook University.
RedShred, a leader in AI-powered text analytics and document understanding, has been awarded a $1.8 million Phase II Small Business Technology Transfer Research (STTR) contract by the U.S. Air Force to advance its groundbreaking AIRworthiness Model Automation for cerTification Evaluations (AIRMATE) system.
In partnership with Stony Brook University’s Center of Excellence in Wireless and Information Technology (CEWIT), this project aims to revolutionize the Air Force’s certification processes by streamlining the transformation of certification documents into MBSE-ready digital models.
“This translational award is the latest achievement as Stony Brook University increases the breadth and depth of our innovation and entrepreneurship activities,” said Michael Kinch, CIO, Office of Research and Innovation. “This particular collaboration is exciting as it allows University expertise and resources to address real-world needs for national defense in partnership with the private sector.”
“This month, we’re thrilled to welcome RedShred to their first New York office in the CEWIT Building within the Research & Development Park, fostering even closer collaboration with Stony Brook University,” said CEWIT Center Director Rong Zhao. “Our partnership has grown significantly over the past two years, and with this move, we’re excited to expand our joint efforts beyond the AIRMATE Phase II award.”
“This Phase II award marks a critical step forward in our mission to revolutionize digital transformation across defense operations,” said Jim Kukla, co-founder of Redshred. “Partnering with Stony Brook University, our AIRMATE technology dramatically cuts the time and resources needed to convert legacy documentation into modern MBSE digital models, enabling faster aircraft modifications and recertifications that directly enhance mission readiness for our warfighters.”
The AIRMATE system addresses the Air Force’s critical need for more efficient, cost-effective certification workflows. By leveraging generative AI and RedShred’s advanced document AI platform, AIRMATE aims to accelerate certification timelines, reduce manual workloads, and improve transparency in military aircraft certification processes. CEWIT and Stony Brook University’s Strategic Partnership for Industrial Resurgence (SPIR) will support graduate students working on the Phase II research project.
“CEWIT is proud to partner with RedShred in bringing cutting-edge digital transformation and AI to the U.S. Air Force,” said Manoj D. Mahajan, senior research scientist at CEWIT. “After securing the Phase I STTR in August 2024, we were thrilled to advance to Phase II alongside RedShred.”
Mahajan, a former federal program manager, serves as the principal investigator (PI) on the STTR grant, working to align university research with industry innovation. Co-PI Pawel Polak, from Stony Brook’s Department of Applied Mathematics and Statistics and an expert in machine learning, brings more than a decade of experience in statistical learning, signal processing and optimization to the project.
“The AIRMATE project merges advanced AI with practical applications. At CEWIT, we utilize our expertise in fine-tuning large language models, reinforcement learning, and retrieval-augmented generation to enhance the system’s performance. These techniques allow AIRMATE to extract and connect certification data from documents, decreasing manual effort and supporting the Air Force’s adoption of model-based systems engineering,” added Polak. “We value the collaboration with RedShred on this initiative, as it enables us to develop innovative techniques and incorporate them into a functional product.”
CEWIT will contribute to the AIRMATE system by developing advanced AI algorithms for:
Automatic extraction of digital certification artifacts from legacy documents Elimination of manual document searches Domain adaptability without requiring large datasets or specialized expertise Seamless integration with popular MBSE tools Cross-referencing capabilities to link related data points
“Looking ahead to Phase III, we’re excited to scale our solution along with the team for wider deployment across defense platforms,” Kukla added. “As we validate our technology & process with Phase I and II, we aim to deliver a commercially viable product in Phase III to address the critical challenges of modernizing technical data throughout the defense industrial base.”
“RedShred has been an active supporter of CEWIT, providing speakers for our conferences, sponsoring our popular Hack@CEWIT student hackathon, and contributing to Stony Brook University,” emphasized Zhao. “And we look forward to many more years of impactful collaboration and innovation together.”
Read more on SBU News.