--- title: What Kills Apps?, Critical Thinking & AI, SwiftUI Blend... description: A roundup episode covering several Swift and AI development topics: critical thinking when using AI coding assistants (avoiding 'cognitive surrender' and... canonical: https://daily.dev/posts/what-kills-apps-critical-thinking-ai-swiftui-blend-modes-context-rot-more-idrtz1gcj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: What Kills Apps?, Critical Thinking & AI, SwiftUI Blend Modes, Context Rot & More | daily.dev og:description: A roundup episode covering several Swift and AI development topics: critical thinking when using AI coding assistants (avoiding 'cognitive surrender' and... og:url: https://daily.dev/posts/what-kills-apps-critical-thinking-ai-swiftui-blend-modes-context-rot-more-idrtz1gcj og:image: https://api.daily.dev/og/posts/idrtZ1gCJ.png og:image:alt: What Kills Apps?, Critical Thinking & AI, SwiftUI Blend Modes, Context Rot & More 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. # What Kills Apps?, Critical Thinking & AI, SwiftUI Blend Modes, Context Rot & More **[Sean Allen](https://daily.dev/sources/seanallen)** · 13 min read · 0 upvotes · 0 comments ## Summary A roundup episode covering several Swift and AI development topics: critical thinking when using AI coding assistants (avoiding 'cognitive surrender' and blindly trusting AI-generated code), why apps that get 'Sherlocked' by Apple's built-in features often survive by going deep for a niche audience rather than broad like Apple's version, a visual guide to SwiftUI's 21 blend modes with practical examples like multiply and color burn, using AI agents to improve in-app copy based on a WWDC session, and an explainer on AI context windows and context rot, including practical mitigation strategies like starting new threads at natural boundaries and maintaining durable source-of-truth files. The episode is sponsored by Bitrig, an AI-powered native Swift app builder. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is context rot in AI coding assistants and why does performance degrade as the context window fills up? Context rot happens when an AI's context window fills with tokens and becomes noisy with old terminal output, build logs, search results, and outdated implementation details, making it harder for the model to find relevant information, similar to finding one sentence in a 2,000-page binder versus a five-page document. Repeated compaction cycles can also lose important information since the AI decides what to keep. _Developers debugging flaky AI agent output can find grounded context-management explainers like this via daily.dev._ ### What does it mean for an app to get 'Sherlocked' by Apple, and does it always kill the app? Getting Sherlocked means Apple builds a competing feature directly into iOS, as it did with background sounds in iOS 15 competing with the ambient-sound app Dark Noise. It does not necessarily kill the app: nearly five years later Dark Noise was still making about 5,000 dollars a month, because Apple's system features must serve everyone and stay basic, leaving room for niche apps to go deep for a specific audience. _Indie developers weighing platform risk against niche-market strategy can track stories like this on daily.dev._ ### How can I prevent AI coding agents from losing important information after multiple context compactions? Start new conversation threads at natural boundaries, such as beginning a new feature or reaching a good stopping point, rather than letting one thread compact repeatedly. Also maintain durable sources of truth as separate files, like an agents.md file, competitor research, or project requirements, so critical information persists outside the AI's memory instead of depending on what compaction chooses to keep. _Anyone tuning long-running AI coding sessions can find practical context-management tactics like these on daily.dev._ ## Similar posts on daily.dev - [Issue 744 – iOS Dev Weekly](https://daily.dev/posts/issue-744-ios-dev-weekly-heaqgwxyz) · iOS Dev Weekly · 1 upvotes · 0 comments - [From Learning to Building: My Journey Through iOS, AI, and Xcode](https://daily.dev/posts/from-learning-to-building-my-journey-through-ios-ai-and-xcode-qelidctof) · Medium · 0 upvotes · 0 comments - [Copyright Walls Demolished by Vibe Coding, and the Developer's New Moat -- Fatbobman's Swift Weekly \#131](https://daily.dev/posts/copyright-walls-demolished-by-vibe-coding-and-the-developer-s-new-moat----fatbobman-s-swift-weekly--rlbbtmi1z) · Fatbobman · 3 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#ios](https://daily.dev/tags/ios), [#ai-coding](https://daily.dev/tags/ai-coding), [#swiftui](https://daily.dev/tags/swiftui), [#indie-hacking](https://daily.dev/tags/indie-hacking) [View this post on daily.dev](https://daily.dev/posts/what-kills-apps-critical-thinking-ai-swiftui-blend-modes-context-rot-more-idrtz1gcj) ```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":"What Kills Apps?, Critical Thinking & AI, SwiftUI Blend Modes, Context Rot & More","url":"https://daily.dev/posts/what-kills-apps-critical-thinking-ai-swiftui-blend-modes-context-rot-more-idrtz1gcj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/what-kills-apps-critical-thinking-ai-swiftui-blend-modes-context-rot-more-idrtz1gcj"},"datePublished":"2026-08-24T14:32:53.240Z","dateModified":"2026-09-13T21:24:12.336Z","description":"A roundup episode covering several Swift and AI development topics: critical thinking when using AI coding assistants (avoiding 'cognitive surrender' and...","image":"https://i.ytimg.com/vi/jxmQPfkwaXo/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/jxmQPfkwaXo/sddefault.jpg","isAccessibleForFree":true,"articleSection":"Sean Allen","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":"Sean Allen","logo":"https://media.daily.dev/image/upload/s--lhkFEwaF--/f_auto/v1722586475/logos/seanallen","url":"https://daily.dev/sources/seanallen"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/what-kills-apps-critical-thinking-ai-swiftui-blend-modes-context-rot-more-idrtz1gcj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,ios,ai-coding,swiftui,indie-hacking","timeRequired":"PT13M","video":{"@type":"VideoObject","name":"What Kills Apps?, Critical Thinking & AI, SwiftUI Blend Modes, Context Rot & More","description":"A roundup episode covering several Swift and AI development topics: critical thinking when using AI coding assistants (avoiding 'cognitive surrender' and...","thumbnailUrl":"https://i.ytimg.com/vi/jxmQPfkwaXo/sddefault.jpg","uploadDate":"2026-08-24T14:32:53.240Z","duration":"PT13M","url":"https://api.daily.dev/r/idrtZ1gCJ","embedUrl":"https://www.youtube.com/embed/jxmQPfkwaXo"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Sean Allen","item":"https://daily.dev/sources/seanallen"},{"@type":"ListItem","position":3,"name":"What Kills Apps?, Critical Thinking & AI, SwiftUI Blend Modes, Context Rot & More"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/what-kills-apps-critical-thinking-ai-swiftui-blend-modes-context-rot-more-idrtz1gcj#faq","mainEntity":[{"@type":"Question","name":"What is context rot in AI coding assistants and why does performance degrade as the context window fills up?","acceptedAnswer":{"@type":"Answer","text":"Context rot happens when an AI's context window fills with tokens and becomes noisy with old terminal output, build logs, search results, and outdated implementation details, making it harder for the model to find relevant information, similar to finding one sentence in a 2,000-page binder versus a five-page document. Repeated compaction cycles can also lose important information since the AI decides what to keep. Developers debugging flaky AI agent output can find grounded context-management explainers like this via daily.dev."}},{"@type":"Question","name":"What does it mean for an app to get 'Sherlocked' by Apple, and does it always kill the app?","acceptedAnswer":{"@type":"Answer","text":"Getting Sherlocked means Apple builds a competing feature directly into iOS, as it did with background sounds in iOS 15 competing with the ambient-sound app Dark Noise. It does not necessarily kill the app: nearly five years later Dark Noise was still making about 5,000 dollars a month, because Apple's system features must serve everyone and stay basic, leaving room for niche apps to go deep for a specific audience. Indie developers weighing platform risk against niche-market strategy can track stories like this on daily.dev."}},{"@type":"Question","name":"How can I prevent AI coding agents from losing important information after multiple context compactions?","acceptedAnswer":{"@type":"Answer","text":"Start new conversation threads at natural boundaries, such as beginning a new feature or reaching a good stopping point, rather than letting one thread compact repeatedly. Also maintain durable sources of truth as separate files, like an agents.md file, competitor research, or project requirements, so critical information persists outside the AI's memory instead of depending on what compaction chooses to keep. Anyone tuning long-running AI coding sessions can find practical context-management tactics like these on daily.dev."}}]} ```