GitHub’s AI-powered Copilot was made available in June this year. GitHub investigated and quantified the impact on developer productivity, finding that 87% of developers found Copilot preserved their mental energy. 74% experienced a rise in overall daily workflow satisfaction.
GitHub Copilot is the artificial intelligence-powered pair programmer that makes helpful suggestions by detecting patterns and prompting with suggestions to finish the code based on history and context. It first went into the technical preview in June 2021 before going live for all developers 12 months later.
The research covers the time from June 2021. Along with the two findings above, GitHub research also found that 73% of developers said it helped them stay in the flow. Dr. Eirini Kalliamvakou conducted the research. More than 50% of the respondents are professional developers.
GitHub also divided a group of 95 developers into two groups to determine the speed of completing a task with and without Copilot. The findings were that those using Copilot completed the task 55% faster than those without, and that is for those who completed the task at all. Those using Copilot had a higher percentage of completing the task, at 78% vs. 70% without.
GitHub had launched Copilot, believing it would improve developer productivity, aid efficiency, and reduce decision fatigue. The research proves the giant developer hub and software repository to be correct, with developers reporting that Copilot helped them stay focused on the task at hand, make meaningful progress, and feel good at the end of a day’s work. Developers reported a significant difference in satisfaction and productivity.
GitHub’s research used the SPACE framework to measure Satisfaction and well-being, Performance, Activity, Communication and collaboration, and Efficiency and flow (hence S-P-A-C-E) and says more insights are still to be released.