ASU students win big at homeland security design challenge

By Cynthia Gerber

Arizona State University students took home five prizes — including two first-place victories — from this year’s Designing Actionable Solutions for a Secure Homeland student design challenge.

Over the course of 48 hours, 160 students from 15 universities tackled some of the most pressing AI-driven challenges facing the Department of Homeland Security, competing for a top spot and a share of the $50,000 prize pool. Participants selected one of three problem statements, each presenting a unique AI-driven security challenge, with first-, second- and third-place awards given for each. 

Dozens of ASU students entered DASSH 2025: Security in an AI World, racing against the clock to develop groundbreaking solutions. The competition challenged students to think beyond code, leveraging AI, engineering and creative problem-solving to address real-world threats.

The weekendlong design challenge was co-hosted by ASU’s Center for Accelerating Operational Efficiency and the Soft Target Engineering to Neutralize the Threat Reality at Northeastern University, or SENTRY, both DHS Centers of Excellence. The largest turnout in DASSH history brought together the country’s brightest student minds for a weekend of caffeine-fueled problem-solving and late-night brainstorming.

"This is not just another academic challenge. It is an opportunity for all of you to design, build and implement innovative solutions that future generations of homeland security practitioners can use," said Julie Brewer, acting undersecretary for the DHS Science and Technology Directorate. "Be inventive and push yourselves in pursuit of solutions to the challenges. In this event, what you are doing really matters; the ideas, concepts and solutions you will put forward this weekend matter.”

“We want to see how you can apply what you’ve been learning to real-world problems, and we want to expose you to academics and government professionals working in these areas. Also, we hope you develop an interest in these fields — and maybe even consider working for DHS,” added Ron Askin, executive director of the Center for Accelerating Operational Efficiency.

The student competitors, many of whom compete in hackathons regularly, saw DASSH as an opportunity to expand their skill sets, collaborate with peers and engage with industry, academic and government leaders.

For ASU physics and engineering student Gavin Russo, a member of ASU’s Extended Warranty team and one of last year’s DASSH winners, the competition provided an unparalleled level of problem-solving and systems-thinking experience to tackle real-world issues.

"Events like these push us to explore ideas we wouldn't have considered in a classroom setting," said Russo. "We’re not just finding the right answer — we’re creating it."

Mentors played a crucial role in guiding students as they developed and refined their projects, offering strategic advice, technical insights and real-world perspectives.

"I encouraged them to share their ideas as succinctly as possible — not only because they had just five minutes to present but also to ensure they had a clear plan," said mentor Kevin Kapadia, a two-time DASSH competitor and former second-place winner, now a PhD student in quantitative psychology at the University of Southern California. "The key to mentoring is getting the participants to talk about their projects, because they usually realize what they are missing just from discussing it."

For many mentors, the challenge wasn’t just about offering technical guidance but also about helping students focus their efforts. "These are big problems," said Dan McCarville, a professor of practice at ASU and longtime mentor. "Students often try to solve the whole thing. Getting them to narrow their scope to something manageable and have a meaningful result in the limited time span is often needed."

Participants worked diligently over the weekend, consulting with mentors and preparing video presentations showcasing their solutions. On the final day, exhausted but exhilarated, teams presented their ideas to a judging panel of homeland security experts, academic leaders and industry professionals.

“The teams delivered solutions very different from one another, even within the same problem statement; some focused on prototyping, others on thorough research, and some proposed bold ideas that, with refinement, could become real-world applications,” said Petar Jevtic, one of the judges and an associate professor in ASU’s School of Mathematical and Statistical Sciences. “The passion of the younger generation towards finding solutions that will be good for our citizens was the most rewarding part of this competition.”

The judges were so impressed with the presentations that they added a fourth place and honorable mention to the prizes available for one of the problem statements.

The following are the winning teams for each of the problem statements:

Problem statement 1: How can we establish a broader comfort with AI to enhance its usage by DHS personnel and in society at large?

First place ($10,000) — Quartz (ASU)
Second place ($5,000) — NSF Seton Hall University (Seton Hall)
Third place ($2,500) — Manual Intelligence (ASU)

Quartz, a team from ASU, won first place with Clear AI, a system designed to enhance trust and transparency in AI-driven decision-making. Their AI model combines deep learning with logical reasoning, offering step-by-step explanations for its outputs — an essential feature for DHS personnel who require accuracy and accountability.

"Unlike traditional AI models, Clear shows its reasoning process clearly rather than relying purely on probability," said Tyler Santos, an ASU computer systems engineering student. "This builds user confidence and fosters AI adoption."

Clear AI reduces errors, enhances security and operates at a fraction of the cost of traditional AI models. The team demonstrated their prototype via a live demo, allowing attendees to test the system in real time using a QR code.

Problem statement 2: How can DHS leverage AI, particularly large language models, to improve efficiency in its operations?

First place ($10,000) — KPS (ASU)
Second place ($5,000) — Team BEC (City College of New York & Rutgers) 
Third place ($2,500) — UB GenAI Mavericks (University at Buffalo) 
Tied for fourth ($100 per team member) — Golden Bulls (Johnson C. Smith University) and Extended Warranty (ASU)
Honorable Mention — MoneyWell (ASU)

KPS, an ASU team, designed Guardian AI, a drone-powered network for automating infrastructure inspections. The AI-driven drone swarms use computer vision and machine learning to detect structural vulnerabilities in real time.

"A traditional DHS inspection might take weeks, but with Guardian, an operator can orchestrate a full scan in minutes," said ASU computer science student Kasyap Rachuri.

During their live demo, the team simulated an oil rig inspection in which Guardian identified a gas leak — a potentially catastrophic failure, also noting Guardian significantly reduces cost, risk and inspection times by up to 80%.

Problem statement 3: How do we mitigate the potential for AI-enabled attacks on critical infrastructure?

First place ($10,000) — DASSHing Ducks (Stevens Institute of Technology)
Second place ($5,000) — Captain AmerIca (University of the District of Columbia)
Third place ($2,500) — Cyber Psyche (ASU)

The DASSHing Ducks from the Stevens Institute of Technology developed LiDAR LASSO, an AI-powered system to neutralize autonomous vehicle threats. Unlike traditional roadblocks, LiDAR LASSO specifically disrupts unmanned vehicles while allowing human drivers to pass freely.

"AI-driven attacks require AI-driven defenses," the team noted. "Our system spoofs LiDAR sensors, tricking autonomous vehicles into stopping while leaving human-driven traffic unaffected."

This low-cost, scalable security measure can be deployed at critical infrastructure sites, high-security areas and public venues.

Unlike typical hackathons, which primarily focus on programming and software development, DASSH emphasizes real-world problem-solving, welcoming students from all disciplines. In-depth AI knowledge was not required for DASSH 2025; instead, the challenge encouraged creative, interdisciplinary solutions to pressing homeland security issues. Participants collaborated with peers, academic experts and industry professionals to develop actionable ideas — whether through AI applications, systems frameworks, engineering solutions or other innovative approaches.

The Center for Accelerating Operational Efficiency is within ASU’s Global Security Initiative and has hosted a student design challenge for years, each time identifying and tackling a new pressing issue, partnering with SENTRY to co-host the most recent competitions. Beyond the competition, DASSH serves as a launchpad for students’ futures.

Competing in their first-ever hackathon-style event, ASU’s four-member Cyber Psyche team, all first-year students, secured third place in DASSH 2025. Their AI-driven phishing detection system demonstrated how students with no prior competition experience could develop impactful solutions.

"Once our families found out we were competing in a three-day design challenge, we were getting supportive WhatsApp messages throughout the weekend," Cyber Psyche team member and ASU computer science student Hrishita Nangunoori said. “Everyone was really excited for us.”

To expand opportunities for students across the country to engage in real-world homeland security problem-solving, the DASSH Student Design Challenge includes a request for proposals initiative for minority-serving institutions. Sponsored by the the Center for Accelerating Operational Efficiency, the RFP provided funding to support faculty and students' participation in the event.

This year’s recipients included Baris Aksanli (San Diego State University), Mohamed Elbakary (Elizabeth City State University), Cotina Lane Pixley (University of the District of Columbia), Suryadip Chakraborty (Johnson C. Smith University) and Patrick Flanigan (North Carolina Central University). These faculty members played a pivotal role in guiding their student teams through the competition, exposing them to real-world problem-solving and career opportunities within the homeland security enterprise.

“Our students have knowledge of AI and machine learning, but many had never seen its applications in homeland security before. This event showed them that AI has a crucial role beyond traditional engineering and science fields,” Aksanli said. “DASSH not only helps students develop expertise in homeland security topics but also provides networking opportunities with industry professionals and other participants.”

“We are always looking for ways to enhance our students’ learning experiences and introduce them to new opportunities. Even for those who don’t pursue a career in AI or programming, DASSH helped develop important skills like time management, decision-making and teamwork,” Flanigan said. “The students told me they had never considered some of the issues they encountered during the challenge, which really broadened their perspective.”