Enablement, spend controls and tool pilots: 17 sourced coding agent stories
Company-level moves: preinstalled or self-hosted setups, peer and champion adoption, budgets and token policies, and pilots or proofs of concept of the tool before wider rollout. This technique sits in the Organisation stage. Of the 17 stories, 8 were published by a tool vendor, 8 by the people who did the work and 1 by independent researchers or press. 10 involve an existing codebase and 1 started from scratch. Every number links to where it was published.
Ticksly, a three-person concert ticketing startup, has agents write all its code: 2,540 merge requests in 28 days at about $8 each
Ticksly · Claude Code, OpenAI Codex
CaveatsWritten by the team that did the work, in a post launching their own open-source tool. Counts merge requests opened, not merged. No figures on quality, defects or revenue. Budgets were added after one week when subagents on an expensive model added $1,900 to the bill.
2,540 opened in the last 28 days
Merge requests
Cisco DevNet's 12-engineer team ships a mandatory PubHub rewrite in Go in about five and a half months instead of nine, using Claude, Copilot and Cursor
Cisco (DevNet PubHub team) · Claude, GitHub Copilot, Cursor
CaveatsWritten by the team itself. Compared against their own pre-AI estimate, not a measured baseline. Claude only joined in June, so the first two months ran at the old pace. No cost, PR or line counts. Engineering detail is promised in a separate AI for SDLC series.
Nine planned months shipped in about five and a half (39% reduction against the pre-AI estimate)
Schedule
Comarch compresses a five-year rebuild of its 10-million-line Clarion ERP to 18 months with Devin
Comarch · Devin
CaveatsPublished by Cognition as a customer story. The 18-month timeline is a plan in progress, not a finished migration. No cost or output code-size figures.
About 10 million lines of Clarion, 20 years old
Codebase
Zalando's 2.5 years of agentic engineering across 250+ teams: a risk-based bot auto-approves 33% of PRs and cuts their lead time 20-40%
Zalando · GitHub Copilot, Claude Code, OpenCode
CaveatsWritten by Zalando's own engineering leadership. The 20-40% lead-time cut is for auto-approved PRs compared with all PRs. No productivity or cost totals; complexity findings come from four example codebases. Also mentions pi, a coding agent not in the site's tool list.
More than 250 engineering teams
Teams
DoorDash's in-house Flux cloud agent platform automates 130,000 engineering tasks in a month
DoorDash · Flux (DoorDash), Agent Gateway (DoorDash)
CaveatsPlatform post from the builders. Which coding agent harnesses Flux runs is not named. Counts are usage volumes; no merge rates, quality or time-saved figures are given.
130,000 in a single month
Engineering tasks automated
Assembled auto-approves low-risk PRs with Codex and Claude Code reviewers, lifting PRs merged to 2.4× its pre-agent baseline
Assembled · OpenAI Codex, Claude Code, Devin, 143.dev (Assembled)
CaveatsSelf-reported by Assembled's CTO. The author calls these observational rollout results, not a randomized trial, and says the auto-reviewer doesn't get credit for every point. The window is short (three weeks after full-team launch), and the bug figures are explicitly too early to be conclusive. The pre-agent baseline period is not precisely dated, and no absolute PR counts are given. Category multipliers differ between the intro (bugfixes 3.7×, refactors 7.5×, features 2.5×) and the results list (refactoring 7.5×, test work 3.4×). The $3.57 per attempt is an estimate, not total spend. Codex and Claude Code are the reviewers; Devin is mentioned only for non-engineer PRs, and the agents that wrote most of the PRs are not broken down.
2.4× pre-agent baseline
PRs merged
Microsoft study of its early-2026 Claude Code and Copilot CLI rollout: adopters merged about 24% more pull requests
Microsoft · Claude Code, GitHub Copilot
CaveatsWritten by Microsoft researchers, and Microsoft owns GitHub Copilot, so labelled vendor. The authors note a merged PR is not the same as the value it delivers.
Tens of thousands of Microsoft engineers
Population
Uber caps AI coding spend at $1,500 per employee per tool after burning its annual AI budget in 4 months
Uber · Claude Code, Cursor
CaveatsPress report based on Bloomberg and The Information. Uber's COO said it is 'very hard to draw a line' between AI usage and new consumer features.
$1,500 per employee, per tool
Monthly cap
Salesforce moves its whole engineering org to Claude Code and says a 231-person-day API migration took 13 days
Salesforce · Claude Code, AI Expert Suite, Salesforce Foundation Plugins, Engineering 360
CaveatsSalesforce writing about its own engineering org; The Decoder notes none of the numbers can be independently verified. The 231 person-days is the team's own estimate (seven per API). The 13-day migration came out as five PRs, the largest covering 21 endpoints with 100% test coverage. No cost or token figures.
33 endpoints in 13 days vs ~231 person-days estimated (18x)
API migration
Intercom lets an AI agent approve 19% of pull requests, with 93% of PRs now agent-driven
Intercom · Claude
CaveatsSelf-reported by Intercom. The post names Claude as writing the code but not the exact tool or the models behind the review agent. Revert-rate comparison is observational: AI-approved PRs are filtered to be small, which may favour them.
Over 93% (two main codebases)
Agent-driven PRs
Vercel auto-merges 58% of PRs in its largest monorepo without human review, cutting average merge time from 29 hours to 10.9
Vercel · Vercel Agent, LLM PR risk classifier (Vercel), Vercel AI Gateway
CaveatsVendor-published: Vercel sells Vercel Agent and AI Gateway, which run the workflow. Dead original link: the original vercel.com address (https://vercel.com/blog/58-percent-of-prs-in-our-largest-monorepo-merge-without-human-review) now returns 404 and is not in Vercel's sitemap, so this story links to the Wayback Machine copy of 7 April 2026 (checked 10 November 2026); all numbers were verified from that archived copy, which shows no author names. The source covers an LLM classifier plus auto-approval, not a coding agent writing the code: it does not say which agents (if any) authored the PRs, and it does not name the classifier model (secondary write-ups say Gemini). Self-reported, from an experiment window of unstated length, in a repo dominated by marketing, docs and UI changes, so transfer to riskier codebases is unproven. The control group had 2 reverts out of 513 reviewed low-risk PRs. Cost figures are per-assessment rates, not a total.
58%
PRs merged without a human reviewer
Dropbox indexes a 550,000-file monorepo in Cursor and accepts more than a million agent-generated lines a month
Dropbox · Cursor
CaveatsVendor-published Cursor customer story. Exact PR-throughput and cycle-time numbers are not given (only “upper tier of industry benchmarks”). Lines “accepted” are not the same as merged production code.
550,000+ files
Monorepo indexed
Uber's uReview AI code reviewer covers over 90% of ~65,000 weekly diffs, with 75% of comments rated useful
Uber · uReview (Uber)
CaveatsSelf-reported by Uber. This is an AI code reviewer, not an agent that writes code, and the post gives no merge-rate, cycle-time or defect outcomes. It does not say which coding agents authored the reviewed diffs, so the tool is Unstated. Published August 2025, earlier than Uber's other stories on the site. Volume figures are inconsistent: '~65,000' diffs is called weekly in one place and 'per month' in another, and 'over 90%' coverage sits awkwardly with 'over 10,000 commits' processed a week. The 1,500 hours a week is an estimate assuming a second human reviewer needs 10 minutes per commit. The usefulness rate counts only engineers who rate comments. Fixer, the companion tool that proposes code changes, is not covered.
Uber Engineering blog (Sonal Mahajan, Shauvik Roy Choudhary, Will Bond)
Over 90% of ~65,000 weekly diffs landed
Coverage
Stripe rolls Claude Code out to 1,370 engineers with a zero-configuration binary
Stripe · Claude Code
CaveatsVendor-published. Stripe states it hasn't isolated metrics attributable to any single AI assistant.
1,370
Engineers with Claude Code enabled
DTCC reports 40% higher developer throughput in an Amazon Q Developer proof of concept
DTCC · Amazon Q Developer
CaveatsVendor-published; results are from the proof-of-concept phase. Build-failure rates reportedly did not increase.
+40% average
Developer throughput (POC)
Palo Alto Networks onboards 2,000 developers to a Sourcegraph Cody + Claude coding assistant in 3 months
Palo Alto Networks · Sourcegraph Cody
CaveatsVendor-published (AWS with partners Anthropic and Sourcegraph). Productivity measurement method not described.
2,000 in 3 months
Developers onboarded
Satispay modernizes its core payment service from Java 8 to Java 21 and Spring in under four days with Claude Code, against a four-week estimate
Satispay · Claude Code
CaveatsAnthropic customer marketing page. No publish date on the page. No independent verification of the four-day migration, defect rates, or the 75% code-share metric. CTO quotes selected by the vendor.
Under 4 days vs 4-week estimate (Java 8 → 21 + Spring)
Core payment modernization