Software developers show strong interest in reducing wasted computing work but often lack practical tools to identify and measure it, according to a GitHub and Yale survey of 1,039 GitHub users in the United States. Eight in 10 respondents wanted tools that help them write more energy-efficient code, and almost three-quarters wanted ways to quantify the impact of software or development practices.
The findings connect routine engineering decisions with growing concern about AI infrastructure. Seventy-one percent of respondents said they were worried about AI systems’ energy use, water consumption and carbon emissions. Seventy-nine percent expressed concern about global warming, while 75% said employers should actively reduce their environmental impact.
Those opinions do not measure the actual footprint of a model, repository or application. They show demand for workflows that can find unnecessary computation, propose an optimization and test whether it helped before maintainers decide to ship it. Reliable measurements would also help teams distinguish meaningful gains from claims based only on cleaner-looking code.
The sample came from US GitHub users who had opted into marketing communications, so it should not be treated as representative of every developer or even every GitHub user. The clearest result is narrower: within this group, interest in efficiency is well ahead of the tools available to support it.