AI and productivity: A year-in-review with Microsoft, Google, and GitHub researchers

40 snips

Dec 29, 2025

Guest

Eirini Kalliamvakou

Join Brian Houck from Microsoft, Ciera Jaspan and Collin Green from Google, and Eirini Kalliamvakou from GitHub as they dive into the complexities of measuring AI's impact on software development. They discuss how traditional metrics like lines of code fall short, the potential for AI to reshape developer roles, and the importance of identifying genuine pain points for AI solutions. The conversation also explores how AI can foster creativity and accelerate junior engineers' growth, setting the stage for a more thoughtful approach to AI adoption in the industry.

Ask episode

AI Snips

Chapters

Transcript

Episode notes

INSIGHT

Measure Developer Experience Holistically

Developer experience requires multiple complementary metrics, not a single productivity number.
Brian Houck recommends measuring speed, efficiency, collaboration, and meaning together to avoid misleading signals.

INSIGHT

Watch Long-Term Trade-Offs

AI introduces trade-offs like de-skilling and re-skilling that unfold over years.
Collin Green urges measuring long-term sustainability and developer growth alongside short-term gains.

ADVICE

Avoid Lines-Of-Code Metrics

Avoid using lines of code as a productivity metric; it misleads more than it informs.
Brian Houck warns that verbosity from AI can inflate LOC without reflecting real value.

Get the Snipd Podcast app to discover more snips from this episode

Get the app