Wayfair rolled out Devin to 1,600+ engineers and merged 40,000+ PRs
Wayfair, the home-goods retailer, moved its engineers onto Devin cloud agents after a bake-off of several platforms. A pilot planned for 20 engineers grew to 100 people and then to company-wide access; Cognition reports that more than 1,600 engineers now use Devin and have merged more than 40,000 PRs with it, with a two-week order-replay test cut to about an hour and a long-stalled database migration moving 8-12x faster.
The problem
Wayfair's engineers already had AI coding tools, but they ran on each developer's own machine, so each person could only work on 1-2 tasks at a time, and they could not easily reach the private, self-hosted systems Wayfair's code depends on. Some work was simply slow: the fulfillment optimization team spent about two weeks per release gathering data and replaying past orders through new routing logic, and a move of core databases off on-premises SQL Server, in instances of up to 30 TB, had been stalled for four years. Engineers had no standard setup and traded AI tool configurations in Slack.
How they did it
- Run a bake-off before you standardize. Wayfair opened several cloud agent platforms to engineers, collected their feedback and judged them mainly on how easily each could reach internal, self-hosted systems.
- Put your hardest problem in the pilot. Cognition asked Wayfair to bring its toughest work, so the stalled database migration was part of the pilot from the start.
- Build one shared environment instead of per-engineer VMs. In Devin, ask a session to "Set up your environment for this repository" (or add repos under Settings > Environment > Blueprints), approve the proposed blueprint, and every session then boots from the resulting snapshot with dependencies, tools and secrets ready.
- Map teams before opening it to everyone. On Devin Enterprise, create organizations per team, connect SSO such as Okta with identity-provider groups, and commit your coding rules in AGENTS.md or similar rules files, which Devin pulls in automatically, so each team's sessions start with its own repos and conventions.
- Let Devin test before a human reviews. Each session runs on its own machine, so have it build, run and test the change; for mobile apps, Devin's docs cover Android emulator and macOS VMs with the iOS Simulator.
- Add automations for recurring work. Under Automations, create an automation with a Slack message, webhook or schedule trigger and a Start session action; Wayfair uses them for incident response, end-to-end and integration testing, ticket triage and backlog grooming.
- Set spending controls early. Use Devin's usage policies (budgets for shared pools, limits per user) and leave model choice to Devin's routing rather than asking engineers to pick models.
From the official docs
Set up your environment for this repository.
Results as reported by Cognition
- Wayfair planned for 500 engineers; more than 1,600 now use Devin and have merged more than 40,000 PRs with it, as reported by Cognition.
- A new engineer can log in and ship a first PR within an hour or two, after Wayfair mapped more than 200 small teams into Devin, as reported by Cognition.
- Replaying past orders through new fulfillment logic went from about two weeks to about an hour, as reported by Cognition.
- In the six-month planning cycle after adopting Devin, the fulfillment optimization software team finished everything it had committed to, took on almost 30% more work, and raised complexity-weighted PR throughput 2.78x, as reported by Cognition.
- The database migration that had stalled for four years is now moving 8-12x faster; the first stage, 17 to 20 databases, was done or nearly done by mid-September, as reported by Cognition.
- About 20% of Wayfair's Devin use runs through automations, and automated testing is up 9.5x for teams using them, as reported by Cognition.
These figures are as reported by Cognition in Cognition (Devin) customer story ↗ (Nov 9, 2026). Cognition sells the product used here. AgentGid did not measure or independently verify them.
This suits large engineering organizations with many internal systems, an identity provider and a platform team that can own the rollout; organizations, IdP groups and budgets for shared pools are Enterprise features, and Wayfair also had a forward deployed engineer from Cognition. A small team can still copy the cheap parts: one shared environment, rules in AGENTS.md and a scheduled automation for tests or triage. Per AgentGid's Devin page (pricing checked Oct 6, 2026), Devin Cloud access starts with Pro at $20 a month (the Free plan covers desktop agents only), Max is $200 a month, Teams is $80 a month plus $40 a month per full dev seat for up to 200 users, Enterprise is custom, and extra usage is bought at API pricing. All results are self-reported in a customer story on Cognition's own site, not independent measurements. For an open-source route, OpenHands is free for local deployment, with cloud Teams at $500 a month for up to 10 seats per its AgentGid page (checked Oct 28, 2026).
The agent used here
Similar use cases
How Nubank split a 6-million-line ETL monolith with Devin
Nubank had Devin carry out the repetitive sub-tasks of breaking its ETL monolith into sub-modules, with engineers reviewing every change. Cognition reports large gains in speed an…
As reported by Cognition: an '8x engineering time efficiency gain' and '20x cost savings'.
How RV Tech sped up vehicle software test generation with Devin
The Rivian–Volkswagen joint venture uses Devin to generate software-in-the-loop tests from requirements and to pre-triage vehicle access tickets. As reported by Cognition, test ou…
As reported by Cognition: a '10-15x increase in test generation velocity'.
GitHub rewrote its Copilot agent runtime in Rust with Copilot CLI
GitHub moved the agent runtime behind Copilot CLI, the Copilot app and the Copilot SDK from TypeScript on Node.js to Rust, with agents in the Copilot app and Copilot CLI writing m…
More than 800,000 lines of production Rust written, mostly by agents, across 128 pull requests that landed in main, as reported by GitHub.
Guides
AI Coding Agent Pricing Explained: Claude Code, Cursor, Codex and Copilot (2026)
What Claude Code, Cursor, OpenAI Codex, GitHub Copilot and other coding agents really cost: plan prices, usage limits, subscription vs API billing, and how to estimate your monthly bill.
Best AI Agents in 2026: Live Rankings by Category
The leading AI agents for coding, design, research, data, marketing, sales, support, HR, finance, legal, DevOps and security, ranked by public usage data that refreshes daily, with prices and free options.
Best AI Coding Agents in 2026: Ranked by Usage, Benchmarks and Price
The AI coding agents developers use most, with live download and install numbers, benchmark results, entry prices and a way to choose between terminal, IDE and cloud agents.