--- title: Practicing engineering judgment with AI | daily.dev description: Practicing engineering judgment with AI means pausing before accepting a coding agent's first patch to decide if it's the right change. Using a hypothetical... canonical: https://daily.dev/posts/practicing-engineering-judgment-with-ai-obtfb0scy twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Practicing engineering judgment with AI | daily.dev og:description: Practicing engineering judgment with AI means pausing before accepting a coding agent's first patch to decide if it's the right change. Using a hypothetical... og:url: https://daily.dev/posts/practicing-engineering-judgment-with-ai-obtfb0scy og:image: https://api.daily.dev/og/posts/ObTfb0sCy.png og:image:alt: Practicing engineering judgment with AI 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. # Practicing engineering judgment with AI **[Flavio Copes](https://daily.dev/sources/flaviocopes)** · 13 min read · 0 upvotes · 0 comments ## Summary Practicing engineering judgment with AI means pausing before accepting a coding agent's first patch to decide if it's the right change. Using a hypothetical CSV export timeout as a worked example, the approach separates observed symptoms from assumed causes, asks agents for evidence rather than diagnoses, and compares design options like streaming versus a background job by the obligations each creates. It also covers writing predictions before reading a patch, choosing tests that could reject the implementation, reading code as a future maintainer, and keeping a short decision journal after shipping. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How can I avoid accepting a bad fix from an AI coding agent without realizing it? Separate the observed symptom from your explanation before asking the agent to fix anything, then give it a bounded investigation task to gather evidence rather than jump straight to a diagnosis. For example, ask it to reproduce a failure and report where time or memory is spent, instead of telling it to 'fix the queue problem,' which biases it toward that explanation. _Engineers refining how they delegate fixes to coding agents follow practical workflow breakdowns like this on daily.dev._ ### What should I test to make sure an AI-generated CSV export fix actually works? Use a small dataset whose expected result you can calculate independently of the exporter's own logic, including records from another account and outside the date filter to catch permission or filter bugs. Also test quoted and multiline CSV fields with a real parser, the original large input that caused the failure, and behavior when a download fails or is canceled partway through. _Developers writing test suites around AI-assisted patches can find similar review checklists on daily.dev._ ### Why might a streaming CSV export fix pass memory tests but still fail in production? Streaming can reduce memory usage by producing output progressively, but it does not guarantee the request finishes within the server's request deadline. In one worked example, measurements supported that whole-file buffering caused memory pressure, yet after switching to streaming the download still hit the request deadline, showing that fixing one constraint can leave another unresolved. _Teams debugging timeout versus memory tradeoffs in long-running exports track these patterns on daily.dev._ ## Similar posts on daily.dev - [How Much Should Engineering Change for AI Agents? A QA Perspective](https://daily.dev/posts/how-much-should-engineering-change-for-ai-agents-a-qa-perspective-usnmsnafd) · Medium · 1 upvotes · 0 comments - [How to Evaluate AI Code Quality: A Practical Guide for Engineers](https://daily.dev/posts/how-to-evaluate-ai-code-quality-a-practical-guide-for-engineers-rbtcmfynl) · freeCodeCamp · 0 upvotes · 0 comments - [The Engineering Manager](https://daily.dev/posts/the-engineering-manager-tvlyneaho) · Tech Lead Digest · 0 upvotes · 1 comments - [Managing the Hidden Overhead of AI Software Engineering](https://daily.dev/posts/managing-the-hidden-overhead-of-ai-software-engineering-odnh0uxvl) · DEV · 1 upvotes · 0 comments - [Software Engineering Principles Help Get More out of AI](https://daily.dev/posts/software-engineering-principles-help-get-more-out-of-ai-g2d3tqsex) · Telerik · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#testing](https://daily.dev/tags/testing), [#code-review](https://daily.dev/tags/code-review) [View this post on daily.dev](https://daily.dev/posts/practicing-engineering-judgment-with-ai-obtfb0scy) ```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":"Practicing engineering judgment with AI","url":"https://daily.dev/posts/practicing-engineering-judgment-with-ai-obtfb0scy","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/practicing-engineering-judgment-with-ai-obtfb0scy"},"datePublished":"2026-10-28T14:26:58.211Z","dateModified":"2026-10-28T15:02:32.739Z","description":"Practicing engineering judgment with AI means pausing before accepting a coding agent's first patch to decide if it's the right change. Using a hypothetical...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a28310be0ec9dba48cf36fa4b2920ebf?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/a28310be0ec9dba48cf36fa4b2920ebf?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Flavio Copes","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":"Flavio Copes","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/1f2dbd275797411bbd37bfdaaf0951a9","url":"https://daily.dev/sources/flaviocopes"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/practicing-engineering-judgment-with-ai-obtfb0scy","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,testing,code-review","timeRequired":"PT13M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Flavio Copes","item":"https://daily.dev/sources/flaviocopes"},{"@type":"ListItem","position":3,"name":"Practicing engineering judgment with AI"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/practicing-engineering-judgment-with-ai-obtfb0scy#faq","mainEntity":[{"@type":"Question","name":"How can I avoid accepting a bad fix from an AI coding agent without realizing it?","acceptedAnswer":{"@type":"Answer","text":"Separate the observed symptom from your explanation before asking the agent to fix anything, then give it a bounded investigation task to gather evidence rather than jump straight to a diagnosis. For example, ask it to reproduce a failure and report where time or memory is spent, instead of telling it to 'fix the queue problem,' which biases it toward that explanation. Engineers refining how they delegate fixes to coding agents follow practical workflow breakdowns like this on daily.dev."}},{"@type":"Question","name":"What should I test to make sure an AI-generated CSV export fix actually works?","acceptedAnswer":{"@type":"Answer","text":"Use a small dataset whose expected result you can calculate independently of the exporter's own logic, including records from another account and outside the date filter to catch permission or filter bugs. Also test quoted and multiline CSV fields with a real parser, the original large input that caused the failure, and behavior when a download fails or is canceled partway through. Developers writing test suites around AI-assisted patches can find similar review checklists on daily.dev."}},{"@type":"Question","name":"Why might a streaming CSV export fix pass memory tests but still fail in production?","acceptedAnswer":{"@type":"Answer","text":"Streaming can reduce memory usage by producing output progressively, but it does not guarantee the request finishes within the server's request deadline. In one worked example, measurements supported that whole-file buffering caused memory pressure, yet after switching to streaming the download still hit the request deadline, showing that fixing one constraint can leave another unresolved. Teams debugging timeout versus memory tradeoffs in long-running exports track these patterns on daily.dev."}}]} ```