Follow ZDNET: Add us as your preferred source Google.
Important points of ZDNET
- Managing technical debt can consume up to 40% of IT development time.
- One way to overcome traditional challenges is to use specialized AI agents.
- Focus on testing tools, refining projects, and driving long-term change.
Businesses are held back by legacy systems. IDC research reports that unmanaged technical debt can consume 20% to 40% of IT development time and divert resources from innovation and modernization.
IDC suggests that while many companies are enthusiastic about deploying AI-enabled services, their ambitions are constrained by technical debt such as outdated systems, weak integrations, and limited data interoperability.
Related article: 5 ways to grow your business with AI – without leaving your employees on the sidelines
The good news is that pioneering business owners are meeting this challenge head-on. While traditional IT burdens can prevent organizations from adopting new data and AI services, some companies are taking a radical approach and using AI to modernize their systems and create new opportunities for in-house development teams.
That’s certainly true, says Jeff Love, chief technology officer of the Professional Rodeo Cowboys Association (PRCA), the sport’s governing body that sanctions rodeo events in the United States, Canada and Mexico.
Mr. Love was passionate about exploring how AI could help the nearly 100-year-old organization overcome the challenges of legacy IT. Here, he offers five lessons for other business leaders looking to take a similar approach.
1. Test your AI model
Love explained to ZDNET how much of the PRCA’s backend systems ran on 40-year-old AS/400 code.
This reliance on legacy systems means development teams spend more time maintaining old code than building new functionality, preventing organizations from embracing digitalization and new ways of working.
“My goal here is to modernize applications, because they’re becoming harder to maintain and we’re losing a lot of knowledge about how to maintain these systems as they get older,” he said.
Related article: 5 ways rules and regulations can guide AI innovation
Love realized that AI could help the PRCA overcome traditional challenges. However, initial tests using a year-old generation AI model yielded mixed results.
“We tried using ChatGPT, but the only difficulty we encountered was the amount of code. ChatGPT couldn’t handle the amount of data we were trying to give it. There were probably close to 1,000 files that we were trying to summarize,” he said.
“Then I read about Grok and thought it might be able to handle some parts of the code a little better. But I tried, and I couldn’t. I tried some other tools that claimed to be able to document the code base, but they weren’t looking at all the files holistically. They were looking at each file and documenting that data.”
2. Use professional solutions
After initially experimenting with AI models, Love started using Zencoder, an agent platform that analyzes business logic and translates it into plain English explanations, last July.
Love: “My goal here is to modernize the application.”
PRCA
Love said the platform sounds like a dream come true for organizations eager to reduce technical debt.
Related article: 5 ways Lenovo’s AI strategy delivers real results
He believed that Zencoder could help PRCA overcome the challenges of legacy code, while traditional approaches and generative AI models were unable to penetrate decades of business logic.
“I tried Zencoder,” he said. “I gave it the AS/400 code and said, ‘Document this, give it the business rules, give it what database files to access, tell it how it can be modernized and what to consider.'”
Love said the initial results were promising, but not perfect. “It wasn’t perfect at first, just because there was a huge amount of information that we had to go through.”
But as his team refined the agent’s work, the PRCA business analyst realized there were tools available to help the organization migrate away from the AS/400 system. “So we started creating more detailed requirements.”
3. Put theory into practice
Love and colleagues gave agents instructions, guidelines, diagrams, and workflows. These key requirements helped create PRCA’s Wiki for Business Analysts.
The organization then created a wireframe based on key requirements and business rules.
“Based on those wireframes, we were able to take the work items and put them into the agent that we created to help with the coding.Then the workflows were brought in and built into the UI structure that we were leveraging in our modernization, and we used that as a starting point for the coding,” he said.
Related article: 5 ways to prevent your AI strategy from breaking down
Love said Zencoder technology has helped staff understand the interconnected nature of code and systems.
Then, as the platform generated new code and modernized legacy systems, unit tests were created to prevent bugs before production.
“We were able to input requirements and set test acceptance criteria to ensure we captured the business rules, so when we modernized the system, we still kept the real rules in mind,” he said.
4. Shift from legacy thinking
PRCA’s technology team is currently modernizing. With the help of AI, time previously spent on legacy issues can be redirected to digitization.
Love estimated that Zencoder reduced development time by 50%, allowing IT teams to build digital services, create new event management tools, and deliver better experiences.
“We don’t have a lot of resources,” he says. “Our small in-house team has six large systems to manage, and the support work required to maintain functionality can be overwhelming.”
Love said that because of the complexity of rodeo’s business logic, it can take a long time for new employees to learn the rules of the sport.
Related article: This company’s AI success was built on 5 key steps – see how they work
Zencoder takes the heavy lifting out of your processes, allowing your staff to work faster and focus on the changes that bring the most value to your organization.
“We can bring in developers and they can work from day one. They understand the logic better, so they can understand what the actual business rules are, so they’re not as afraid to make changes,” he said.
“Right now, we spend a lot of time doing things like unit testing, all of which are essential to building robust applications, but unfortunately, we can end up putting these tasks on the back burner as we focus on features and product releases.”
5. Find new challenges
Love said his team aims to complete the organization’s AS/400 system migration by the end of 2026.
Once this work is complete, the team will move on to the next legacy platform, PRCA’s ASP.NET Web Forms technology.
“The first goal is to get ourselves up to speed,” he said. “We’re 40 years old. Once we graduate from AS/400, it’s 20 years old. Our next big project will be moving from ASP.NET to more modern applications.”
Love said the long-term goal is for the team’s agent-facing processes to help the organization continue to evolve digitally.
“We are now in the second year of our five-year plan to bring us into the modern era,” he said. “But at the end of that period, there will be other projects that were started in the first place, and there will be a time to start on them, refresh them, and deal with new business rules.”
