Computing student tackles ‘technical debt’ with prestigious Department of Energy award

Eric Melin, a third-year doctoral student in the computing program, is one of 79 doctoral students from across the U.S. selected for the Department of Energy's prestigious Office of Science Graduate Student Research Program

For Melin, this award not only validates the critical nature and promise of his research, but it also empowers his patriotic sensibilities and desire to use his skills to serve his country.

Of his program placement at Oak Ridge National Laboratory, Melin said "receiving this award is a tremendous honor. I'm excited to work on research that I can be truly proud of and that has real-world impact. The experience has been incredibly rewarding. I've had the chance to work alongside brilliant scientists on fascinating problems, and I'm gaining a real sense of what it's like to work at a national lab. Having the opportunity to contribute to the DOE is especially meaningful to me, because I've always wanted to serve the country I call home. My mother came here as an immigrant, and I feel a strong sense of responsibility to give back."

In his research, Melin applies AI and data science techniques to analyze scientific software, managing technical debt to improve its quality, maintainability and reliability. "Scientific software" is software that domain scientists write to complete specific scientific and research-related functions.

A key challenge with scientific software is that the scientists programming it may - or may not - have a formal background in software development. Combined with the pressure to advance research and meet strict deadlines, this may lead scientists to create shortcuts in their software to expedite results.

These shortcuts are referred to as "technical debt" and could prove expensive, time-intensive and a poor use of resources in the long run.

IBM defines technical debt as "the future costs associated with relying on shortcuts or suboptimal decisions made during software development. Also called code debt or design debt, these compromises are primarily due to quick fixes, poor documentation and reliance on outdated code. Over time, this debt must be addressed, requiring additional effort. This 'repayment' typically involves refactoring, debugging and ongoing code maintenance."

Fortunately, Melin is already making headway addressing this technical debt with his method.

"I recently finished my first proposed project, which focuses on automatically identifying and classifying different types of technical debt across several software artifacts," Melin said. "By extracting text from selected DOE code repositories and applying my trained classifier, I can detect instances where debt is being admitted, helping practitioners to remediate and correct these shortcomings in the future."

Melin is thankful for the support he has received along the way.

"I am very grateful for my mentors and my time at Boise State which has shaped me into the computer scientist and researcher I am today," he said. "A special thanks to Nasir Eisty for encouraging me to apply for this fellowship and the countless others that supported me in the application process. I am extremely proud to represent the School of Computing and Boise State with this award."

Share This

• Share on Facebook
• Share on X
• Share on LinkedIn
• Share through Email