Can AI approve code changes without a person?

“The change sets the tier, not the author. A classifier reads the change and picks its risk tier. … An agent's approval never counts as a person's, either. Automated review can clear a low-risk change, but it can't fill a slot that calls for a human.”

AnthropicPublished by the organization itself (post), no named author

“The builder verifies with targeted checks and runs a quick self-review. CI runs the deterministic harness. The deep review runs on request in an independent session, with the verdict returned to whoever asked. The author's watch loop handles threads and formal re-requests. Humans own the required approvals, and bot or agent verdicts do not satisfy them.”

TradersPostPublished by the organization itself (post), no named author

“You get a pretty high degree of confidence on merge. Today, almost a fifth of all the pull … are merged without any human review or without any human testing. … that is a number that we care a lot about and want to bring higher and higher, of course, within the guardrails of producing really high quality code.”

Daksh GuptaCo-founder · Executive, Greptile

“Auto-approve is not auto-merge. When Diff Vader approves, it posts an APPROVE review, the same thing a human reviewer clicking “approve” does. It does not push the merge button. The engineer merges (or GitHub’s native auto-merge does, if they enabled it), and every existing gate—CI, branch protection, required-reviewer counts—still applies.”

Dave NorthLeads the technical operations group · Leader, Rewind

“So we built a system that fixes the allocation. It’s called Diff Vader: an AI reviewer that looks at every PR across our orgs and, for a carefully bounded set of low-risk changes, becomes the only reviewer needed, freeing human reviewers to spend their judgment where judgment actually matters.”

Dave NorthLeads the technical operations group · Leader, Rewind

“It declares the hard rules the verdict engine is not allowed to override no matter what the council says: paths that touch authentication, authorization, secrets, encryption, and anything that changes who is allowed to approve future changes always require a human.”

Dave NorthLeads the technical operations group · Leader, Rewind

“we had code owners in our code base and they were kind of right most of the time, like 80% of the time. But for these 20% of the time, they cost a lot of bottlenecks for us internally. … So what we started doing was building this agentic code owner thing. And what it essentially does is look in at PRs and check in like, first of all, what's the risk of this? What's the risk level? … And if it is a low risk, it can just approve the PR …”

Eric ZakariassonEngineer, Developer Experience and Product, Cursor · Practitioner, Cursor (Anysphere)

“We started with a controlled pilot of over 100 PRs through the AI approval pipeline. The results: zero reverts of AI-approved PRs, and a 6–16x improvement in time-to-approval at the 75th percentile. Since then, the system has scaled significantly. In the first four weeks of broader rollout, 497 PRs went fully autonomous, with Claude writing the code and our AI approval system reviewing, approving, and shipping to production.”

Intercom (renamed Fin in May 2026)Published by the organization itself (post), no named author

“Today, over 93% of our pull requests (PRs) across our two main codebases are Agent-driven. And over 19% are auto-approved with no human reviewer in the loop.”

Intercom (renamed Fin in May 2026)Published by the organization itself (post), no named author

“The AI agent can approve the review, but a person always clicks the merge button. That merge is the traceable approval event for SOC 2: the agent handles the review workload, the human controls what ships.”

OnaPublished by the organization itself (post), no named author

“First, engineers cannot self-classify a change as low-risk. An Ona Automation evaluates every PR against objective criteria automatically, which removes the temptation to game the system and makes the boundary auditable.”

OnaPublished by the organization itself (post), no named author

“Our engineers waited a median of 2 hours 49 minutes for their first PR approval. Four weeks after we let an AI agent approve low-risk PRs, that dropped to 3.8 minutes, lead time fell 74%, and deploys tripled, all with the same size team.”

OnaPublished by the organization itself (post), no named author

“We run Code Review on nearly every PR at Anthropic. Before, 16% of PRs got substantive review comments. Now 54% do. It won't approve PRs — that's still a human call — but it closes the gap so reviewers can actually cover what's shipping.”

AnthropicPublished by the organization itself (post), no named author

“The AI is approving over 60% of all PRs. For those PRs, engineers get feedback in minutes, not hours.”

Joey BatesDeveloper Enablement · Practitioner, Customer.io

“High-risk PRs that the system flagged for human review took about a day to approve. That's longer than the baseline. This is correct behavior. These are architectural changes, infrastructure modifications, and breaking changes. They need time and human judgment.”

Joey BatesDeveloper Enablement · Practitioner, Customer.io

“We taught AI to approve pull requests for merging. After 500 AI-approved PRs: zero production incidents, and merge time dropped from 3.5 hours to 33 minutes.”

Joey BatesDeveloper Enablement · Practitioner, Customer.io

“right now we still require an engineer to review every PR … goes in. … Once the more confidence we get in our risk scoring and we can also back test it with years of data, right? We have all of this data and the more confidence we have in our AI code reviews, then maybe for some class of low risk PRs, we could enable, we call it green lane like auto landing.”

Uday Kiran MedisettyDistinguished Engineer, Uber · Practitioner, Uber

“The reviewer agent surfaces issues and inconsistencies; the human makes the final call on quality and correctness.”

WarpPublished by the organization itself (documentation), no named author

“So the goal is that most pull requests in the Claude Code repos … merge with at most one human in the loop, who can be the author, with no increase in reverts or incidents. To measure human time, we track human approvals per PR (median and 95th percentile), human comments, and how long a PR waits on a person. Revert and incident rates are the guardrail: if they rise, we haven't saved human time, we've only moved it.”

AnthropicPublished by the organization itself (post), no named author

“We don't yet have circuit breakers that tighten approval when those numbers move; engineers here want them before more code moves to zero-human merges.”

AnthropicPublished by the organization itself (post), no named author

“Currently, we still have humans approving the code, but we see a short path in the near future to a percentage of our code landing automatically, having automatic approvals, right?”

Will BondStaff Engineer, AI Foundations team, Development Platform organization · Practitioner, Uber

“So as I said, doing pull requests … not an ideal situation. So we want to remove ourselves from that step to make it fully autonomous. I do think maybe that we need another like Opus 4.5 moment that LLM's, there's a new version that just is better. … But there are a lot of changes that could go straight to test. Like the document search could have gone without us approving it. But … it kind of needs to build this track record of doing stuff right before we trust it. So we're not quite there yet. But I think it's a matter of time.”

Hallstein BrøtanSenior consultant, Novanet; developer on Nordic Corporate Bank's system · Practitioner, Nordic Corporate Bank

“cortex went all-in: the AI writes the code, the AI reviews the code. What humans keep their hands on is "tuning the prompts and guidelines themselves" -- not making decisions inside each individual PR, but watching the system from above and adjusting.”

Ryosuke TsujiCTO · Executive, airCloset

“Every one of those 769 PRs had an AI reviewer as the first reviewer, with an average of 10.8 review-fix loop iterations per PR (max 56). 1 in 5 merged within 10 minutes, roughly half within 30 minutes. What humans do now is look at review outcomes and tune the review prompt and the guidelines themselves -- this is human-on-the-loop, not human-in-the-loop. Humans operate on the policy layer, not the execution layer.”

Ryosuke TsujiCTO · Executive, airCloset

“we are allowing the bot to merge that directly into our main branch because we have integration tests, unit tests, etc., canary tests.”

Xiucheng QuekSoftware engineer · Practitioner, Genentech

“Our Agent is strict. It won’t approve large PRs. If a change is too big, too complex, or too broad in scope, it flags it and requires it to be broken down. Creating a direct, positive incentive for engineers to ship smaller, more incremental, well-scoped changes.”

Intercom (renamed Fin in May 2026)Published by the organization itself (post), no named author

“The bias is explicitly toward approval, meaning a single warning in an otherwise clean MR still gets approved_with_comments rather than a block.”

Ryan SkidmoreAuthor of Cloudflare's "Orchestrating AI Code Review at scale" blog post · Practitioner, Cloudflare

“That is fast enough that most engineers see the review comment before they have finished context-switching to another task. The metric we’re the proudest about, though, is that engineers have only needed to “break glass” 288 times (0.6% of merge requests).”

Ryan SkidmoreAuthor of Cloudflare's "Orchestrating AI Code Review at scale" blog post · Practitioner, Cloudflare

“We've been running this system internally across tens of thousands of merge requests. It approves clean code, flags real bugs with impressive accuracy, and actively blocks merges when it finds genuine, serious problems or security vulnerabilities.”

Ryan SkidmoreAuthor of Cloudflare's "Orchestrating AI Code Review at scale" blog post · Practitioner, Cloudflare

“Most AI code review tools stop at comments. They suggest improvements, flag issues, and point out problems. We built a system that approves PRs for merging.”

Joey BatesDeveloper Enablement · Practitioner, Customer.io

“we've merged more than 2.5 million automated maintenance PRs, the vast majority auto-merged with no human in the loop.”

SpotifyPublished by the organization itself (post), no named author