Spotify says its top developers haven’t written a line of code since December, thanks to artificial intelligence

Has AI coding reached a tipping point? At least that appears to be the case for Spotify, which shared this week during its Q4 earnings call that the company’s top developers “haven’t written a single line of code since December.” This statement from Spotify co-CEO Gustav Soderström came alongside other comments about how the company is using artificial intelligence to speed up the development process.

It’s worth noting that Spotify noted that it shipped more than 50 new features and changes to its streaming app throughout 2025. Recently, it rolled out more features, like AI-powered requested playlists, page matching for audiobooks, and About This Song information, all of which have been rolled out over the past few weeks.

At Spotify, engineers are using an internal system called “Honk” to speed up the coding process and product speed, the company told analysts on the call. This system allows for things like remote, real-time code deployment using generative AI, specifically Claude Code.

“As a concrete example, an engineer at Spotify during his morning commute from Slack on his cell phone could ask Claude to fix a bug or add a new feature to the iOS app,” Soderstrom said. “Once Claude is done with that work, the engineer gets a new version of the app, sent to them on Slack on their phone, so he can then integrate it into production, all before they get to the office.”

Spotify credited the system with helping to speed up the coding and publishing process “tremendously.”

“We don’t expect this to be the end of AI development, just the beginning,” Soderstrom said.

The CEO also praised Spotify’s ability to build a unique data set that other MBAs can’t commoditize, in the same way they can use other online resources, like Wikipedia. That’s because there isn’t always a realistic answer to questions about music, he said.

For example, if you ask what workout music is, you’ll get different answers from different people, sometimes based on their geographic location. Americans tend to prefer hip-hop music in general, although millions prefer metal. While a number of Europeans are into electronic dance music, many Scandinavians love heavy metal.

“This is a dataset that we’re building now that no one else is building,” Soderstrom noted. “It doesn’t exist at this scale. And we see it getting better every time we retrain our models.”

Analysts on the call also asked about Spotify’s approach to AI-generated music. The company explained that it allows artists and labels to indicate in a track’s metadata how the song was created but still monitors the platform for spam.