John Shelburne bought something valuable and realized he did not have the keys. He had decades of market experience, but suddenly he held a complicated codebase and a responsibility he could not outsource. That moment forced a decision: learn quickly and thoroughly, or watch potential slip away. What followed wasn’t a smooth journey. It was hands-on, sometimes messy, and intensely practical. It required asking uncomfortable questions, making avoidable mistakes, and rebuilding confidence one technical step at a time.
A Hidden Treasure with a Steep Price to Learn
John had about 25 years in financial markets, beginning with a decade at Bloomberg and then moving into sales and trading. In 2021, he put in a modest bid for the intellectual property of Katana Labs, a fintech tool that used machine learning to detect anomalies in the bond market. Against the odds, he won the bid. He later described the outcome simply as acquiring a product that was worth between ÂŁ1.5 million and ÂŁ2 million for just ÂŁ25,000. The win was real. The challenge that followed was immediate.
“What I acquired was not a simple app but a deep, multi-branch repository, a Cartesian product of market relationships encoded in data pipelines, machine learning models, and deployment scripts,” he said. “I had domain expertise but no developer experience. I was completely in over my head.”
Recognizing that gap, John set out to learn deliberately. He enrolled in the Professional Certificate in Full Stack Software Development: Building Scalable Cloud Applications by the McCombs School of Business at The University of Texas at Austin to gain the practical technical knowledge he needed. The program was a means to an end, not the headline of the story. The headline was the work he had to do to make sense of what he now owned.
Building Systems That Matter
The moment John opened GitLab for the first time, the scale of what he had acquired hit him. “When I first looked at it, I had no idea what a repo was,” he said. “I saw it for the first time and thought, I can handle those couple of folders. There were probably about 20 branches inside it. And within those branches, there were several projects.” That’s when he started to understand the gravity of what he had acquired.
He realised that to unlock the real value of the product, he needed more than intuition, he required technical skill. “To gain value out of what I had acquired, my motivation was that I needed to go back to school and learn what I actually had. And what I had was a machine learning tool that identifies anomalies in the bond market and helps you find pair trades.”
This is where the Professional Certificate in Full Stack Software Development: Building Scalable Cloud Applications became crucial. The program gave him the structure and clarity to understand the repository, trace component interactions, and start contributing to the code with confidence. With that foundation, John applied his learning directly at Catfix Technologies, the company he founded to deliver data-driven financial solutions. He moved code across clouds, standardised environments, and inspected Docker images to understand exactly what was running in production. From there, he refined orchestration, reduced costs, and improved reliability.
“It was not just about learning new technology. It was about optimising the systems I was already working with, which saved me both time and money,” he said. The technical skillset enabled him to turn a complex inherited codebase into a product that could be improved, scaled, and handed off with clarity.
Learning by Doing, One Bug at a Time
John’s learning was never abstract. From the first lessons, he pushed straight into the technical areas that mattered for his codebase. “If you know the process, it is much easier to move forward in software development,” he said. Version control became a foundation. Understanding Git, branches, unit testing, and a clear lifecycle turned confusing code interactions into a manageable process.
He worked through front-end and back-end concepts, but his focus was always practical: how to find the bug, how to trace the pipeline, how to deploy without breaking production. He learned what Docker images meant, how a container differed from a virtual machine, and why infrastructure as code mattered. He began to recognize the languages used across the repo and to ask targeted questions in code reviews rather than being baffled by jargon. “Learning by doing made all the difference. The ability to take what I learned and immediately apply it to my own project gave me both the confidence and the competence to move forward,” he explained.
John made mistakes but didn’t back down. They marked learning milestones. He also learned to frame problems differently. Instead of asking how to make some piece of code run, he began to ask what business requirement that code was serving.
Practical lessons and a clear playbook
John’s advice for any learner is simple. “The basics are everything. If you do not understand how things work at a fundamental level, you will never be able to solve the complex problems that come later,” he said. He also emphasized patience and questions. “The best piece of advice I have is to be patient and ask a lot of questions,” he told interviewers. That approach shaped how he moved from domain expert to someone who could read and influence code.
John’s journey was never about an easy transition. He had something valuable in his hands and realized that to manage it well, he needed to build new skills from the ground up. He went from understanding markets to understanding systems. He went from inheriting a codebase to creating clear processes around it. In the end, he gained not just technical skill but the ability to make smart, business-minded decisions about the product’s future. The key lesson from his story is simple: ownership drives focus. When you are responsible for something, learning becomes practical and purposeful.
