Using AI to Improve Code Quality: A Practical Approach


Everyone is searching for the most efficient and effective way to use AI in software development. So how do we take advantage of AI and where is the best place to start?  

AI excels at generating code, but it's not always perfect for what you are looking for.  Ensuring consistency, adhering to code patterns, and maintaining structure requires engineers with experience. Based on my experience, approximately 90% of AI-generated code still needs some modifications or tweaking to align with best practices, scalability, and ongoing maintainability. This is where experienced engineers play a critical role in shaping AI-assisted code into production-ready solutions.



AI as a Strategic Tool in Engineering Teams

Over the past several years, I have worked closely with engineering teams to determine how AI can be used reliably and consistently in the development process. While AI can be leveraged across multiple stages of coding, one of the best starting points is improving code coverage through automated testing.

AI and Automated Code Coverage

AI is an excellent tool for helping teams build out automated test coverage, ensuring code quality and stability. One of the teams I worked with had a requirement of 80% code coverage for all new code check-ins. Initially, meeting this requirement took additional effort—about 50% of our time was spent writing the feature code, while the other 50% was dedicated to writing automated tests.

To improve efficiency, we explored whether AI could help reduce the time required for test creation. Unlike feature code, where AI-generated output can vary significantly, test code follows established patterns, making it easier to train AI to produce consistent results. Over time, we found that AI-assisted test generation reduced our test-writing effort from 50% of total development time to just 30% or even 20% in some cases.

Embracing AI as a Tool for Engineering Excellence

Engineering teams need to embrace AI as a tool—one that, when used correctly, can elevate a good engineer to a great one. The key is to start small and identify areas where AI can handle repetitive, time-consuming tasks. Automated testing is a perfect starting point. With better test coverage, teams can spend less time on maintenance and more time focusing on new feature development and innovation.

AI is not here to replace engineers—it’s here as a tool to help enhance their capabilities. The teams that leverage AI strategically will gain a significant advantage in software quality, efficiency, and overall engineering excellence.  


Comments

Post a Comment

Popular posts from this blog

AI Coding Tools, One Year Later: What’s Really Changed?

Image
As an engineering leader, I am always looking for ways to help my engineering team not only achieve the goals of the organization but really enjoy what they are doing. They love to code, and I want to help them do that. Part of that means taking time to find ways and tools to get the right outcomes. So how can AI help with that? A year or so ago, I set out to see how great the AI tools really were at replacing software engineering efforts. I learned a lot, and they absolutely made me a better engineer, but it still felt like they had a long way to go. Recently, I’ve been hearing from some big hitters in the industry that AI coders are now much closer to replacing engineers. Some claims even say that premium services offer AI coding bots that could actually replace a lot of engineering work. Even though I didn’t have access to those premium services, I decided it was time to jump back in and retest the waters with the newly improved tools. How much better could they really be? My Proje...
Read more

Exploring AI Coding Tools: Can They Really Build Software for You?

Image
TL;DR   AI tools like ChatGPT-4 and GitHub Copilot, in their current state, have incredible potential to boost productivity. While they still require substantial hand-holding, reviewing, and slight alterations at nearly every step, they enabled me to complete my project 75%-80% faster thanks to my coding expertise. By using these tools, I was significantly more capable of getting features done in far less time. Though AI coding companions are promising, they are not yet fully autonomous—engineers remain essential in guiding and refining the work to achieve optimal results. It will be interesting to see how long it takes before these tools evolve to a point where engineers may not be needed for the majority of the development process.  For the full story, continue reading below. Introduction: My Objective  As an engineering leader with 20 years of development experience, I aim to help my engineering teams work faster and more efficiently. To do that, I wanted to explore th...
Read more