AWS Public Sector Blog

University of Arkansas – Pulaski Tech increases campus app engagement by 253% with generative AI from AWS Partner Modo Labs

AWS Branded Background with text "University of Arkansas - Pulaski Tech increases campus app engagement by 253% with generative AI from AWS Partner Modo Labs"

To support a student population that includes many nontraditional learners, such as working adults, caregivers, and first-generation learners, the University of Arkansas – Pulaski Technical College (UA-PTC) needed a faster and more intuitive way to deliver campus information. UA-PTC worked with Modo Labs, an Amazon Web Services (AWS) Partner, to launch MyAgent, a generative artificial intelligence (AI) agent built on AWS that delivers real-time, verified answers to conversational student questions directly within the college’s Modo-powered app.

Within weeks of launch, MyAgent became the app’s most-used feature: Students now use the agent daily to navigate enrollment, financial aid, and academic services, driving a 253 percent year-over-year increase in engagement with admissions content.

In this blog post, learn how Modo Labs worked closely with the AWS Product Acceleration Team to develop and deploy MyAgent in just eight weeks, and how UA-PTC is improving student services with generative AI-powered support.

Identifying a new opportunity to improve student access

UA-PTC is the largest two-year college in central Arkansas, serving more than 6,700 students across 80 degree and certificate programs. Many of those students are balancing jobs, parenting, or both, and a significant portion are first-generation college students navigating higher education systems for the first time.

UA-PTC had already been working with Modo Labs, an education technology (EdTech) company and longtime AWS Partner, to deliver and maintain its mobile and web experiences. During a virtual user group call hosted by Modo, the Modo team demoed a prototype of a generative AI agent that could seamlessly integrate with the existing digital campus experience. UA-PTC immediately recognized the opportunity to simplify access to student resources with the dynamic natural language queries supported by a generative AI chatbot.

“I think there was a lightbulb moment for them when they saw the demo,” recalled Andy Umans, general manager of higher education at Modo. “UA-PTC has a lot of nontraditional learners… they wanted to make it easy for students to get information and get over intimidation factors like the friction of having to search for information, maybe with terminology and such that they’re not familiar with.”

Bringing a prototype to production with the AWS Product Acceleration Team

To move its generative AI prototype into production, Modo turned to its AWS account team, which introduced them to the AWS Product Acceleration Team for dedicated support.

AWS’s Product Acceleration Team is a group that specializes in helping customers rapidly build innovative products using AWS tools.

“AWS put together a lot of really useful information,” said Max Denmat, head of partnerships and innovation at Modo. “They sat with us multiple times to review our architecture and offered a lot of guidance throughout the process to help us bring it to life.” The AWS team helped Modo fine-tune key aspects of the solution, from backend architecture to performance, scalability, and security, which helped Modo move fast.

“One of the key factors in terms of this partnership with AWS was also the accelerated generative AI development they offer through Amazon Bedrock,” explained Harsh Paleja, vice president of product management at Modo. “That was an ‘Aha’ moment for us, to see that we could deploy something out in the market very quickly that resonates with the pain points we’re seeing from our customers and prospects.”

From prototype to campus-wide rollout in just eight weeks

With support from the AWS Product Acceleration team and a focused collaboration with UA-PTC, Modo built and deployed the MyAgent generative AI agent in just eight weeks, rolling out across the UA-PTC campus in March 2025. “I don’t know that we’ve ever shipped a new product that fast,” noted Denmat.

The solution runs entirely on AWS, using Amazon Bedrock with Claude 3 Haiku for its large language models (LLMs), AWS Lambda for backend processing, Amazon Simple Storage Service (Amazon S3) for frontend hosting, and Amazon API Gateway for WebSocket connections.

To promote responsible AI deployment, Modo implemented several safeguards. The AI agent responds only to verified institutional content and redirects users to official resources when answers fall outside scope. “It’s okay for it to say, ‘I don’t know,’ or ‘Here’s where you can find more information’ rather than trying to answer something that may not be accurate,” said Paleja. The team also added canned responses for sensitive topics and used AWS tools to redact personal information like Social Security numbers.

Modo tested the assistant with a group of 20–25 students over two weeks, using their feedback to refine the experience. The streamlined UX and prominent placement in the app’s sticky navigation helped drive adoption immediately after launch.

Transforming student information services with MyAgent

UA-PTC is one of the first community colleges to deploy generative AI at scale in this way, and since going live, the results have been incredibly positive. Within weeks of launch, MyAgent became the app’s most-used feature, accounting for 21.8 percent of all mobile and desktop activity and generating twice as many interactions as the next most-used module. With a 94.5 percent user adoption rate across students, faculty, and staff, it’s clear the tool is meeting a real need.

Beyond day-to-day support, MyAgent is also helping UA-PTC attract and onboard new students. Views to the app’s “Future Student” section rose by 140 percent year over year, and Enrollment & Admissions content saw a 253 percent increase in unique visitors. Financial Aid content engagement also increased, with views of Financial Services up 13 percent in the app and 29 percent on the college website. And it’s not just students benefiting: Student workers have begun using the AI agent at front desks to help answer peer questions, turning MyAgent into trusted guidance for the campus community.

“MyAgent is already transforming how our students access information,” said UA-PTC Chancellor Dr. Summer DeProw. “At UA-PTC, we are committed to improving student success through innovation, and this AI assistant is enabling students to get answers to their questions instantly, any time of day.”

Scaling accessible student support with Modo Labs and AWS

Finding basic campus information shouldn’t be a barrier to student success. Modo Labs’s MyAgent meets students where they are to deliver instant answers when they need them most.

For institutions considering similar solutions, Paleja recommends starting small. “UA-PTC started with a set of real questions that they know people have, and they weren’t afraid to start small with a simple handbook,” he said. “Get to that level of accuracy, get comfortable with that, and then start to build on top of it.”

Could your university benefit from a generative AI assistant feature? Explore how Modo Labs can help colleges and universities provide more unified and equitable access to the digital campus experience.

Learn how AWS helps EdTechs build, deploy, and scale solutions that support student success. Contact us today.

Read related stories on the AWS Public Sector Blog: