River threads jump to 121,804 as Shopify's AI agent floods its Slack

What's the deal? Shopify's internal AI agent River handled 121,804 Slack threads in September 2026, nearly quadrupling from 35,418 in March, chief executive officer Tobi Lütke said in an October 19 post on X. The chart, which Lütke dubbed "River floods Shopify," tracked the surge over six months.

What is River? River is an AI agent that lives inside Shopify's public company Slack, built on Shopify's "Aquifer" platform. Employees @-mention it in open channels, where it reads code, runs tests, opens pull requests, queries the data warehouse, and inspects production traces.

Roughly one in eight of Shopify's merged pull requests is now coauthored by River. The company launched the agent in early 2026.

Why now? Lütke published the figures in direct response to Anthropic's study "How employment is changing at Anthropic," mirroring its scatter-plot visuals. He framed River as a parallel data point from outside the Bay Area AI labs.

Thread volume grew far faster than the number of people involved: Shopify employees appearing in the River graph ("Shopifolk" in Lütke's chart) rose from 4,147 in March to 6,273 in September, while links between people writing in the same River threads grew from 61,680 to 171,853.

What made it possible? River rests on infrastructure bets Shopify placed in 2024: moving to a single monorepo it calls "World" and building everything with Nix for reproducible environments. "Every change we made for agents was also the right thing for humans," the company wrote.

River only works in the open — no direct messages — so every conversation becomes a searchable, public transcript other employees can learn from. That design turns individual agent sessions into shared company context.

What could go wrong? Thread volume alone does not prove productivity gains or code quality. Shopify has not published an independent measure of business impact tied to the reported figures.

The signal: Anthropic's study found interactive AI agent hours outnumbered employee hours six to one by August, as the share of junior technical hires fell from 27% to 9%. Shopify's numbers suggest that same adoption curve is now showing up inside a traditional commerce company, not just a frontier AI lab.

Read more: x.com/tobi, shopify.engineering, anthropic.com