Anthropic’s daily code merges jump 10x since May as Claude takes over the keyboard

Anthropic’s engineers are shipping code at a pace that broke their own plumbing. The company says daily code merges climbed from approximately 2,500 in May 2026 to over 30,000 by September 2026.

That is a tenfold increase in roughly four months.

The surge pushed Anthropic to replace Git with a custom-built system it calls Scion. The company laid out the switch in a detailed engineering blog post published on October 13, 2026.

The numbers behind the surge

The driver is Claude Code, Anthropic’s agentic coding tool.

Earlier 2026 reports indicated that Claude was writing more than 80% of Anthropic’s merged production code. Before February 2025, that share reportedly sat in the single digits.

Engineers were merging about 8x more code each day compared with their pre-2025 baselines.

Claude also autonomously handled maintenance routines, generating 388 pull requests, with roughly 46% of them merged after human review.

Why Git stopped being enough

Once Claude started producing the bulk of the code, the bottleneck shifted. Writing code was no longer the slow part. Source control, builds, and continuous integration (CI) became the new constraints.

Anthropic’s answer was Scion, a homegrown version control system. Alongside it, the company upgraded supporting tools, including a code review system called Antique.

The full migration took five months, completed while merge demand kept climbing.

Demand that has not peaked

Anthropic said latent demand for merging code remains very high. In its framing, clearing any remaining bottleneck would immediately translate into more output.

Background: from assistant to primary author

Before February 2025, Claude’s contribution to Anthropic’s codebase was minimal. By early 2026, reports put Claude’s share above 80% of merged production code. By September, total daily merges had passed 30,000.

The research also places Anthropic within a wider trend among leading AI labs. As coding tasks get automated, operational constraints get redefined, and attention moves toward infrastructure orchestration and efficiency rather than pure coding.

What this means

Headline figures like 8x more code per engineer only matter if the code passes review and lands safely. The roughly 46% human-reviewed merge rate on Claude’s maintenance pull requests shows that human judgment still filters a large share of machine output.

Anthropic sells Claude Code to outside developers, and running the company’s own engineering on it serves as a large internal stress test. A lab that can show its tool driving a tenfold jump in merges, and then document the infrastructure needed to absorb it, has a compelling pitch for enterprise buyers.

Anthropic chose to build Scion rather than stretch Git further. If other AI-heavy engineering teams hit similar walls, the market for version control and CI tooling could face its biggest rethink in years.