GitHub has a front-row seat to how code is changing now that everyone—and their army of agents—can ship code. In March alone, agents created 17 million pull requests on the platform.
That’s why I was thrilled @hammer_mt was on hand to interview @github COO @kdaigle at Microsoft Build for a behind-the-scenes look at how the platform is helping developers manage the influx without dictating which pull requests they should trust or merge.
This week on @every’s AI & I, Mike gets into: - The 14x commit explosion. GitHub hit 1 billion commits last year. Kyle says they’re on pace for 14 billion this year—and he doesn’t think that curve is plateauing. - GitHub is committed to letting open-source maintainers set their own standards. Agent PRs are flooding communities, but GitHub’s philosophy is to leave code maintainers in control. - The developer/non-developer distinction is collapsing. GitHub’s own legal and finance teams are using Copilot to build apps, one example of how AI has expanded the definition of who counts as a developer. - Per-seat pricing doesn’t survive a world where agents run while you sleep. Kyle thinks automatic model routing—swapping in Haiku for simple tasks instead of always calling the expensive model—is the best way to make the economics make sense. - Daigle runs a daily self-improvement loop with an AI he named Baxter. Every day, Baxter reads 7 days of his emails and Slack messages, flags his communication patterns, and checks whether Kyle followed last week’s advice.
This is a must-watch for anyone running agents in their dev workflow—and curious how GitHub is handling the explosion of commits on its platform.
Watch below!
Timestamps: Introduction: 00:00:52 The agentic PR flood: 00:03:27 GitHub’s approach to helping open-source maintainers manage the surge: 00:04:33 What 14 billion commits means for code quality: 00:06:15 Moving from per-seat licensing to usage-based pricing: 00:08:03 Kyle's dual role as GitHub COO and Microsoft's chief marketing officer for developers: 00:09:45 Developer choice as competitive moat: 00:13:03 How to balance dogfooding your own tools with staying honest about the competition: 00:14:57 Hill climbing, frontier tuning, and solving the model-routing problem: 00:19:45 Kyle's agentic communication hack: 00:24:45
