<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit" -->

---
title: Shopify Just Released The Greatest Claude Code Workflow Ever
description: Shopify rebuilt its main mobile app (300 screens) using AI coding agents through a structured internal tool called Helix, which breaks large tasks into ordered...
canonical: https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Shopify Just Released The Greatest Claude Code Workflow Ever | daily.dev
og:description: Shopify rebuilt its main mobile app (300 screens) using AI coding agents through a structured internal tool called Helix, which breaks large tasks into ordered...
og:url: https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit
og:image: https://api.daily.dev/og/posts/rq3kYZJIT.png
og:image:alt: Shopify Just Released The Greatest Claude Code Workflow Ever
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Shopify Just Released The Greatest Claude Code Workflow Ever

**[AI LABS](https://daily.dev/sources/ailabs-393)** · 14 min read · 1 upvotes · 0 comments

## Summary

Shopify rebuilt its main mobile app (300 screens) using AI coding agents through a structured internal tool called Helix, which breaks large tasks into ordered checkpoints reviewed by humans and gated by four checks: a behavior gate (automated tests), a UI gate (design comparison, using Gemini for spatial reasoning), an adversarial code review gate (two agents arguing over code quality), and a final human review. Since Helix is internal and unreleased, a YouTube channel called AI Labs reverse-engineered the workflow, built their own version with an 'orchestrator' skill, sub-agents, a checkpoint planner, a hook that forces agents to keep working until gates pass (borrowed from the 'Ralph loop'), and HTML prototypes for design comparison when no reference app exists. The video includes a sponsored segment for Hedra, an AI media generation MCP tool, and promotes the creators' paid community (AI Labs Pro) where the built skills are sold.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.youtube.com/watch?v=bBMp5tLxShQ>

## Questions this post answers

### What is Shopify's Helix workflow for rebuilding apps with AI coding agents?

Helix is an internal Shopify tool that breaks a large app rebuild into small ordered checkpoints, each of which must pass four gates before the next begins: a behavior gate running automated tests, a UI gate comparing screens against a design reference using Gemini for spatial accuracy, an adversarial code review gate with two agents arguing over code quality, and a final human review. Sub-agents handle most work to keep each context window fresh.

_Developers building similar multi-agent pipelines can track workflow breakdowns like this on daily.dev._

### Why does Shopify use Gemini models specifically for UI design review in AI coding workflows?

Gemini models were chosen as the design reviewer because testing showed they have strong spatial awareness, letting them accurately judge where elements sit on a screen and how large they are. This lets Gemini catch small spacing and sizing differences between two screens when comparing a rebuilt UI against its original design reference.

_Anyone comparing AI models for UI review work can follow model capability discussions on daily.dev._

### How can I stop an AI coding agent from skipping steps or shipping unfinished work?

Use a hook that runs whenever the agent tries to stop and throws an exit code that forces it to keep working until defined gates are passed, rather than relying on prompt-based rules the agent can forget. This concept, borrowed from the Ralph loop, pairs with checkpoints of increasing complexity so early mistakes are caught while still cheap to fix.

_Developers designing reliable agent guardrails can follow workflow techniques like this on daily.dev._

## Similar posts on daily.dev

- [I built a website with Claude Code and Stitch 2.0, and now I understand why developers are switching](https://daily.dev/posts/i-built-a-website-with-claude-code-and-stitch-2-0-and-now-i-understand-why-developers-are-switching-2gbhxwzdr) · XDA Developers · 1 upvotes · 0 comments

---

Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#claude-code](https://daily.dev/tags/claude-code), [#google-gemini](https://daily.dev/tags/google-gemini), [#shopify](https://daily.dev/tags/shopify)

[View this post on daily.dev](https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"Shopify Just Released The Greatest Claude Code Workflow Ever","url":"https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit"},"datePublished":"2026-09-24T15:41:34.920Z","dateModified":"2026-09-24T15:42:31.667Z","description":"Shopify rebuilt its main mobile app (300 screens) using AI coding agents through a structured internal tool called Helix, which breaks large tasks into ordered...","image":"https://i.ytimg.com/vi/bBMp5tLxShQ/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/bBMp5tLxShQ/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI LABS","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI LABS","logo":"https://media.daily.dev/image/upload/s--UJNjhNyP--/f_auto,q_auto/v1769283201/logos/ailabs-393","url":"https://daily.dev/sources/ailabs-393"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,claude-code,google-gemini,shopify","timeRequired":"PT14M","video":{"@type":"VideoObject","name":"Shopify Just Released The Greatest Claude Code Workflow Ever","description":"Shopify rebuilt its main mobile app (300 screens) using AI coding agents through a structured internal tool called Helix, which breaks large tasks into ordered...","thumbnailUrl":"https://i.ytimg.com/vi/bBMp5tLxShQ/sddefault.jpg","uploadDate":"2026-09-24T15:41:34.920Z","duration":"PT14M","url":"https://api.daily.dev/r/rq3kYZJIT","embedUrl":"https://www.youtube.com/embed/bBMp5tLxShQ"}}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI LABS","item":"https://daily.dev/sources/ailabs-393"},{"@type":"ListItem","position":3,"name":"Shopify Just Released The Greatest Claude Code Workflow Ever"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/shopify-just-released-the-greatest-claude-code-workflow-ever-rq3kyzjit#faq","mainEntity":[{"@type":"Question","name":"What is Shopify's Helix workflow for rebuilding apps with AI coding agents?","acceptedAnswer":{"@type":"Answer","text":"Helix is an internal Shopify tool that breaks a large app rebuild into small ordered checkpoints, each of which must pass four gates before the next begins: a behavior gate running automated tests, a UI gate comparing screens against a design reference using Gemini for spatial accuracy, an adversarial code review gate with two agents arguing over code quality, and a final human review. Sub-agents handle most work to keep each context window fresh. Developers building similar multi-agent pipelines can track workflow breakdowns like this on daily.dev."}},{"@type":"Question","name":"Why does Shopify use Gemini models specifically for UI design review in AI coding workflows?","acceptedAnswer":{"@type":"Answer","text":"Gemini models were chosen as the design reviewer because testing showed they have strong spatial awareness, letting them accurately judge where elements sit on a screen and how large they are. This lets Gemini catch small spacing and sizing differences between two screens when comparing a rebuilt UI against its original design reference. Anyone comparing AI models for UI review work can follow model capability discussions on daily.dev."}},{"@type":"Question","name":"How can I stop an AI coding agent from skipping steps or shipping unfinished work?","acceptedAnswer":{"@type":"Answer","text":"Use a hook that runs whenever the agent tries to stop and throws an exit code that forces it to keep working until defined gates are passed, rather than relying on prompt-based rules the agent can forget. This concept, borrowed from the Ralph loop, pairs with checkpoints of increasing complexity so early mistakes are caught while still cheap to fix. Developers designing reliable agent guardrails can follow workflow techniques like this on daily.dev."}}]}
```

