Skip to content
IonWarpTry for free
AI-written code

Add AI Architecture Review
to your AI-written pull requests.

Flags the duplicate helpers, side paths and leftover code that AI-generated pull requests add, before they merge.

By Ilya Volodarsky · Updated

IonWarp vs AI Reviewers: +8% faster · +69% cheaper
Try for free See the benchmark
IonWarpbotexample review
IonWarp Review — PR #412
3 issues| 2 P1 1 P2
#SevIssueFile
1Second timeout parser reads a bare number as seconds, not millisecondslib/review-timeout.ts
2Import still points to the helper deleted by this PRlib/tasks.ts
3The fix patches one caller; the shared client still drops the errorapi/teams.ts
lib/review-timeout.ts
717273
−+ 
const timeoutMs = parseReviewTimeout(value);const timeoutMs = parseDurationMs(value);// shared parser owns supported units and validation
A second timeout parser reads numbers differently · AI Architecture Review

The new review-only parser duplicates the platform duration parser but treats a bare number as seconds instead of milliseconds. Use the shared parser and delete the duplicate.

Suggested fixAI fix prompt

What AI slop looks like in a codebase, and how to stop it at review

AI slop in AI-written and vibe-coded pull requests rarely looks broken. It looks like a second helper beside one that already exists, a second code path beside the function it should have fixed, a fallback that hides an error, or code the agent left behind.

AI Architecture reviews pull requests for exactly this, and each duplication finding names both copies and the input where they differ. On 11 benchmark pull requests with 6 known issues, DeepSeek V4.1 Flash, the model it runs, found 4; the CodeRabbit app scored highest, an F1 of 72.7% to 47.1%.[1]

The patterns AI Architecture flags

  • A second implementation

    A new parser, client or helper beside the one the codebase already has. The finding names both, and where their behavior differs.

    AI ArchitectureOn Pro and Max

  • A second pathway for one job

    A fast path, a legacy path or a fallback that does the same job a different way: two places to fix every future bug.

    AI ArchitectureOn Pro and Max

  • Code left behind

    Callers, flags, tests and types that only existed for the code this pull request removed.

    AI ArchitectureOn Pro and Max

  • Errors swallowed instead of handled

    Bare catches and failure paths with no log, metric or named error.

    Log & ErrorsOn Pro and Max

AI Architecture Review on the benchmark

ModelF1PrecisionRecall$ per reviewMedian time
DeepSeek V4.1 FlashThe model it runs · best on this benchmark47.1%36.4%66.7%$0.056m 14s
GLM 5.3A failed review retries on it47.1%36.4%66.7%$0.071m 29s

Why IonWarp runs DeepSeek V4.1 Flash: in a trial of our specialist reviewers on production pull requests on 23 September 2026 it filed about as many findings per review as GLM 5.3 Flash at high effort (0.93 against 1.02) for a third of the cost ($0.039 against $0.125), and 3 of its 64 findings were refuted.

Measured on the same 11 open-source pull requests, with known issues. See the benchmark

IonWarp caught 3 issues CodeRabbit, Greptile, Cursor Bugbot or Qodo missed

Real IonWarp comments on open-source pull requests, graded against the benchmark's answer key.

Missed byCodeRabbit 1Greptile 1Cursor Bugbot 0Qodo 3

1 of 3 · See the benchmark

authz/client/src/main/java/org/keycloak/authorization/client/util/crypto/AuthzClientCryptoProvider.javakeycloak #33832
IonWarp caught it Qodo missed it
IonWarpbotcommented on line 114

P2Dead ASN1Encoder calls in concatenatedRSToASN1DER · AI Architecture Review

Lines 114-115 create two ASN1Encoder instances, encode r and s into them and discard the results. The value actually returned is built again at lines 117-121. This is generated filler: it has no effect and the PR should not add it.

LowKnown issue in the AI Architecture Review benchmark · caught by DeepSeek V4.1 FlashSee the graded case

Pricing

Simple pricing for your whole team.

One workspace price with seats included. Credits pool across your team. Unlimited repos on every plan.

Starter

$0 /mo

Try AI PR review.

  • 2 reviewers
    • Code Review
    • Security Review
  • 3 seats included
  • 15,000 credits to start, then 5,000/mo
  • Unlimited repos
Get Started

Pro

Most popular
$49 /mo

Get AI PR review right from day 1.

  • +5 reviewers
    • DevOps Review
    • Docs Review
    • AI Architecture
    • Log & Errors
    • Cost Review
  • 5 seats included
  • +25,000 credits/mo
  • +$29/mo per added seat
  • Unlimited repos
Choose Pro

Max

$149 /mo

Scale AI PR review.

  • +9 reviewers
    • UX Review
    • API Performance
    • Analytics Review
    • SEO Review
    • Changelog
    • CI/CD Performance
    • Plan Review
    • Repeat-Failure Learnings
    • SOC 2 Review
  • 10 seats included
  • +75,000 credits/mo
  • +$29/mo per added seat
  • Unlimited repos
  • Priority support
Choose Max
Enterprise

Custom reviewers + credits · GitHub Enterprise · SSO/SAML + audit logs · Custom security review + SLA

Contact us
FAQ

Frequently asked questions

What is AI slop in code?

Code that works but should not exist in its current shape: duplicate implementations, second pathways for one job, fallbacks that hide errors, and code left behind after a change. Coding agents produce it because they add code fast and rarely delete what is already there.

Can AI review AI-generated code?

Yes, with models that did not write it. IonWarp's reviewers run DeepSeek V4.1 Flash, GPT 5.6 Luna Pro, GLM 5.3 Flash, and they re-review the fix your coding agent pushes.

Will it ask me to add more code?

Rarely. Most findings ask you to delete the second copy and keep one implementation of each thing.

Which plans include the AI Architecture reviewer?

AI Architecture runs on Pro and Max. Pro is $49/mo for the whole workspace with 5 seats included.

Sources

  1. The AI architecture review benchmark: 11 pull requests, 6 known issues (release of 2026-09-24)IonWarp benchmark · read