--- title: Going All In on AI Agents for Five Weeks: What I Learned description: An Artsy engineer describes rebuilding the City Guide product in five weeks using AI agents like Opus 4.5 and Codex. He says orchestration, product management... canonical: https://daily.dev/posts/going-all-in-on-ai-agents-for-five-weeks-what-i-learned-ggysprblb twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Going All In on AI Agents for Five Weeks: What I Learned | daily.dev og:description: An Artsy engineer describes rebuilding the City Guide product in five weeks using AI agents like Opus 4.5 and Codex. He says orchestration, product management... og:url: https://daily.dev/posts/going-all-in-on-ai-agents-for-five-weeks-what-i-learned-ggysprblb og:image: https://api.daily.dev/og/posts/GGYSPRBLb.png og:image:alt: Going All In on AI Agents for Five Weeks: What I Learned 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. # Going All In on AI Agents for Five Weeks: What I Learned **[Artsy.net](https://daily.dev/sources/artsy)** · 12 min read · 0 upvotes · 0 comments ## Summary An Artsy engineer describes rebuilding the City Guide product in five weeks using AI agents like Opus 4.5 and Codex. He says orchestration, product management and QA planning became bigger parts of the job than writing code. Prototypes made it easier to demo and validate ideas, and MCP integrations let agents validate backends, manage tickets and build a Notion knowledge base. He also reports downsides: agents quietly make product decisions, output quality varies day to day, and he's left with 'cognitive debt' from code he didn't fully review. A missed Android QA scenario still reached production, fixed later via Expo Updates. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What problems came up when relying heavily on AI coding agents to build a product quickly? Engineers accumulate 'cognitive debt' where they can't explain why code behaves a certain way without asking an agent first, agents silently make small product and design decisions like date formatting or sorting order, and the same model's output quality can vary noticeably day to day, sometimes requiring a switch to a different model like Codex. _Anyone weighing AI-agent speed against long-term code understanding can track these tradeoffs on daily.dev._ ### How can automated AI code review help a team ship faster with fewer human reviews? An automated AI code review action running on every commit, with a continually improved review prompt, let one team ship faster and with more confidence even when most pull requests weren't reviewed by humans. Documenting recurring mistakes agents make and feeding that back into the review process further improved results over time. _Teams tuning AI-assisted review workflows can follow emerging practices like this on daily.dev._ ### What went wrong when using AI agents for QA testing on a mobile app feature? Agents sped up QA by spinning up browser sessions and simulators and testing flows end to end, but an incomplete QA plan still let bugs through: an Android-specific scenario was missed during testing of a City Guide feature and reached production, later fixed with Expo Updates. Agents execute a QA plan well but don't fix gaps in what gets tested. _Developers relying on AI for QA automation can compare real pitfalls like this on daily.dev._ ## Similar posts on daily.dev - [I Stopped Coding for 5 Years. AI Helped Me Ship Mobile Apps Again](https://daily.dev/posts/i-stopped-coding-for-5-years-ai-helped-me-ship-mobile-apps-again-z52sfqzxh) · ProAndroidDev · 0 upvotes · 0 comments - [AI App Development: What I Learned in One Month](https://daily.dev/posts/ai-app-development-what-i-learned-in-one-month-2tskdg55u) · SwiftLee · 1 upvotes · 0 comments - [How to Perform Effective Project Management with AI](https://daily.dev/posts/how-to-perform-effective-project-management-with-ai-jkp7tkjwc) · Towards Data Science · 1 upvotes · 0 comments - [Building an AI Agent in 6 Weeks \(and Finally Understanding How They Work\)](https://daily.dev/posts/building-an-ai-agent-in-6-weeks-and-finally-understanding-how-they-work--bcuoxcenf) · Planet Python · 17 upvotes · 0 comments - [My First Two Weeks as an Agentic Engineer](https://daily.dev/posts/my-first-two-weeks-as-an-agentic-engineer-ibtc3m6h6) · The Miners · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#mcp](https://daily.dev/tags/mcp), [#claude](https://daily.dev/tags/claude), [#code-review](https://daily.dev/tags/code-review), [#vibe-coding](https://daily.dev/tags/vibe-coding) [View this post on daily.dev](https://daily.dev/posts/going-all-in-on-ai-agents-for-five-weeks-what-i-learned-ggysprblb) ```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":"Going All In on AI Agents for Five Weeks: What I Learned","url":"https://daily.dev/posts/going-all-in-on-ai-agents-for-five-weeks-what-i-learned-ggysprblb","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/going-all-in-on-ai-agents-for-five-weeks-what-i-learned-ggysprblb"},"datePublished":"2026-10-28T10:49:20.319Z","dateModified":"2026-10-28T15:00:07.150Z","description":"An Artsy engineer describes rebuilding the City Guide product in five weeks using AI agents like Opus 4.5 and Codex. He says orchestration, product management...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/d814c0aaacf44d08c95a458addea9c01?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/d814c0aaacf44d08c95a458addea9c01?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Artsy.net","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":"Artsy.net","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/d72060ae77ca4103b2c34329b400a673","url":"https://daily.dev/sources/artsy"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/going-all-in-on-ai-agents-for-five-weeks-what-i-learned-ggysprblb","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,mcp,claude,code-review,vibe-coding","timeRequired":"PT12M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Artsy.net","item":"https://daily.dev/sources/artsy"},{"@type":"ListItem","position":3,"name":"Going All In on AI Agents for Five Weeks: What I Learned"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/going-all-in-on-ai-agents-for-five-weeks-what-i-learned-ggysprblb#faq","mainEntity":[{"@type":"Question","name":"What problems came up when relying heavily on AI coding agents to build a product quickly?","acceptedAnswer":{"@type":"Answer","text":"Engineers accumulate 'cognitive debt' where they can't explain why code behaves a certain way without asking an agent first, agents silently make small product and design decisions like date formatting or sorting order, and the same model's output quality can vary noticeably day to day, sometimes requiring a switch to a different model like Codex. Anyone weighing AI-agent speed against long-term code understanding can track these tradeoffs on daily.dev."}},{"@type":"Question","name":"How can automated AI code review help a team ship faster with fewer human reviews?","acceptedAnswer":{"@type":"Answer","text":"An automated AI code review action running on every commit, with a continually improved review prompt, let one team ship faster and with more confidence even when most pull requests weren't reviewed by humans. Documenting recurring mistakes agents make and feeding that back into the review process further improved results over time. Teams tuning AI-assisted review workflows can follow emerging practices like this on daily.dev."}},{"@type":"Question","name":"What went wrong when using AI agents for QA testing on a mobile app feature?","acceptedAnswer":{"@type":"Answer","text":"Agents sped up QA by spinning up browser sessions and simulators and testing flows end to end, but an incomplete QA plan still let bugs through: an Android-specific scenario was missed during testing of a City Guide feature and reached production, later fixed with Expo Updates. Agents execute a QA plan well but don't fix gaps in what gets tested. Developers relying on AI for QA automation can compare real pitfalls like this on daily.dev."}}]} ```