AWS Public Sector Blog
How California Community Colleges combats fraud with Pariveda Solutions and AWS
An estimated one in eight financial aid applications submitted in higher education may be fraudulent, contributing to billions of dollars in misallocated funds nationally, according to data highlighted by EDUCAUSE.
At California Community Colleges (CCC), the challenge is even more severe: administrators estimate that nearly one in five applications is fraudulent. That creates a difficult choice for the college system, whose core mission is to make higher education as accessible as possible. To intelligently and thoughtfully flag fraud at scale without restricting students’ ability to apply, the CCC turned to Amazon Web Services (AWS) and AWS partner, Pariveda Solutions, to develop a solution.
In this blog post, you will learn how CCC and Pariveda built a machine learning (ML)-powered fraud detection and prevention solution that helps identify and manage fraudulent accounts before they reach the campuses.
Protecting accessible education while combating fraud
California Community Colleges (CCC) is the largest higher education system in the country, serving 2.1 million students across 116 campuses. With 70 percent of their students coming from diverse ethnic backgrounds, the CCC is committed to working with vulnerable populations and making their educational programs accessible to everyone. This nearly frictionless accessibility extends to the application and enrollment process, which creates a fundamental tension in the CCC’s ability to fight fraud effectively.
Even with its robust technical expertise, the CCC Technology Center found itself challenged by the scale and sophistication of thousands of fraud attempts. The CCC Technology Center recognized they needed a custom, internally hosted approach that could evolve as quickly as the threats they faced, while still preserving the open access that defines their mission.
The CCC Technology Center turned to AWS for support. After assessing the scope and complexity of the challenge, AWS recommended working with Pariveda, an AWS partner with deep experience in strategy and ML solutions. Supporting mission-driven engagements across nonprofits, government, and higher education is also a key part of Pariveda’s culture. “It’s become really important to do work that mattered to our team and that felt like it was making a difference in the world,” said Michael Hollar, lead for public sector at Pariveda.
The Pariveda team developed a close collaboration with the CCC Technology Center to learn more about the problem and how they might approach it. According to Sean McKenna, vice president at Pariveda, earning the client’s trust from the start was critical, as Pariveda prioritized transparent communication through each stage of development. This foundation, combined with frequent check-ins and continuous AWS support, created the confidence CCC needed to tackle such a sensitive challenge.
A measured, scalable approach to higher education fraud detection
CCC’s Technology Center took a thoughtful, scientific approach to their fraud prevention strategy, beginning with a proof of concept (POC) designed to validate their initial hypothesis about using ML with historical data.
Following the POC’s success, the team advanced to full development. The resulting solution uses Amazon SageMaker, AWS Step Functions, AWS Lambda, and Amazon DynamoDB. According to Faith-Anne Strieber, a program and technical lead at Pariveda, the solution uses an ML model that is regularly retrained against new data to assess suspected fraud as accounts and applications enter the system. The system-wide approach gives the CCC Technology Center insights that individual colleges couldn’t achieve on their own.
“We’re focused on catching fraud at the very first point of contact,” said Strieber. “This not only reduces downstream risk, but also lessens the burden on campuses that don’t have forensic fraud teams.”
Something that was a significant consideration during this process was data reuse and ensuring the CCC Technology Center stayed within the budget they had set for API requests. With the system processing between 300,000 and one million applications per month, data efficiency was crucial. According to Strieber, this required some creativity to make sure that the multiple parties involved in building this solution weren’t pulling the same data. “It’s about making sure that what we’re building is simple to understand and cost-effective,” she said.
Early results against fraud and a stronger path forward
In one month alone (August 2025), the new solution impacted nearly 20 percent of all applications. The account model caught five percent more fraud than would have been detected without it, while simultaneously allowing 15 percent more legitimate applications to pass through to colleges without friction. This meant institutions received significantly more valid applications while fraudulent submissions were filtered out more effectively—saving time and money for colleges and ensuring genuine students experienced a smoother application process.
In addition to performing well against historical data, the solution hasn’t resulted in any operational disruptions at the CCC. All admissions and records workflows remain unchanged, and while schools see fraud likelihood scores, they still retain full decision authority. “Over time, we hope that trust is built to a point where the schools could say, ‘Hey, if your confidence is that high at account creation, just automatically block that account,” McKenna said.
The Pariveda team praised the CCC Technology Center for their innovative and thoughtful approach to protecting their institution from fraud. “They’re doing things my commercial clients don’t do,” said Hollar. “Their technical aptitude exceeds their reputation, and they’ve taken a strategic, scalable approach to a complex, system-wide problem.”
Moving forward, this fraud detection and prevention solution will save more than just dollars, as the impact of higher education extends far beyond funding. In some cases, fraudulent students have distorted enrollment data, leading colleges to overestimate demand and over-assign faculty. “The faculty thinks they have five sections, when in reality, they have two,” said Hollar. “That trickles down into lost opportunities for both faculty and real students who need those classes.” Halting fraudulent accounts at the first point of contact helps genuine students progress better through their academic careers.
Empowering higher education institutions against fraud
The CCC’s approach proves that with the right partner and a methodical approach, institutions can protect resources while maintaining the open access that defines community colleges.
McKenna’s advice to institutions facing fraud challenges is direct: “One of the things we’ve learned here is, the worst thing you can do is nothing. If you have a fraud problem that’s costing you millions to hundreds of millions of dollars, this is such a low investment to test something out for impact.”
For higher education institutions ready to take action against fraud, Pariveda brings proven expertise in building cost-effective, mission-aligned solutions powered by AWS.
Learn more about how AWS and its partners support colleges and universities in developing secure, scalable solutions that protect access, reduce fraud, and empower institutions to serve students more effectively.
Read related stories on the AWS Public Sector Blog: