Fastest new tool adoption in 25 years is not quite magic, but is getting close…
When engineers at Tampa’s MercuryWorks custom software and .NET development company began a trial of Microsoft’s “AI pair programmer” tool early this year, the firm’s leadership was not particularly enthusiastic.
Founding Partner Donald Bickel described the outlook as a “skeptical view.”
With 25 years of experience specializing in application development and legacy application modernization, firm leaders have witnessed plenty of ambitious product rollouts.
The firm does not see AI tools as a means of generating novel software code – that’s a job for professional engineers. They were more interested in the tool’s capacity to boost productivity and aid quality, as demonstrated over a period of hands-on use.
The results? The quickest new tool adoption the firm had seen in its entire history.
As founder Bickel writes, “We’re not dealing with magic yet, but even this skeptical author has to admit – we’re getting close.”
Here’s an overview of the almost magic MercuryWorks’ engineers found over the course of their trial.
Robust Autocomplete Shrinks Time & Errors
Bickel described the tool’s autocomplete function, which completes entire chunks of code and suggests best practices, as the most used feature within the firm, with day-to-day benefits that included reduced development time, fewer errors, enhanced code discovery, improved code consistency and continuous learning via the introduction of new coding patterns and best practices, according to this source.
Other clear use cases and productivity benefits include:
- Autocompletion and streamlining of repetitive tasks in a way that went much further than the IntelliSense embedded into Microsoft Integrated Development Environments (IDEs)
- Automated unit test creation with streamlined test generation, improved code quality and early bug detection. Copilot also helped catch potential edge cases while suggesting remaining tests that should be created
- Detailed explanations of code sequences that can act as inline documentation within the code itself, reducing the need for separate documentation efforts and enabling faster resolution of issues and streamlined debugging
- Code refactoring suggestions for reduced complexity and improved readability for code improvement, maintainability and reduced risk of regressions. Alternative code structure techniques helped engineers experiment with elegant approaches in a low-risk environment
- Code creation suggestions for increased development speed, improved functionality, risk reduction and enhanced collaboration
- Streamlined debugging process that included suggested solutions. While these suggestions require review and human expertise, by leveraging the capabilities of Copilot, MercuryWorks found, you can streamline the process, identify new solutions and deliver more reliable code
- Automated code documentation with clear and concise explanations of functionality, improving code understanding for everyone working on a project and ensuring a uniform documentation style
Day-to-Day Benefits & Productivity Improvement
In general, Bickel writes, the tool led to solid productivity gains and helped streamline and reduce repetitive tasks. MercuryWorks engineers reported reduced development time, fewer errors, enhanced code discovery and improved consistency, among other benefits, says this source.
The company’s engineering leaders estimated that productivity improved by 5-10% as a result, a gain they expect to grow as Copilot becomes more ingrained into MercuryWorks’ coding style.
“Just by integrating generative AI into the editor, GitHub Copilot has quickly defined a new age of software development, resulting in clear gains of developer productivity and happiness,” Bickel writes.
