--- title: 10 tips to level up your ai-assisted coding - Aleksander... description: A conference talk presenting practical tips for maximizing AI-assisted coding workflows. Key themes include managing context windows carefully (treating them... canonical: https://daily.dev/posts/10-tips-to-level-up-your-ai-assisted-coding---aleksander-stensby---ndc-ai-2026-kjyhgwek6 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: 10 tips to level up your ai-assisted coding - Aleksander Stensby - NDC AI 2026 | daily.dev og:description: A conference talk presenting practical tips for maximizing AI-assisted coding workflows. Key themes include managing context windows carefully (treating them... og:url: https://daily.dev/posts/10-tips-to-level-up-your-ai-assisted-coding---aleksander-stensby---ndc-ai-2026-kjyhgwek6 og:image: https://api.daily.dev/og/posts/kjYhgWEK6.png og:image:alt: 10 tips to level up your ai-assisted coding - Aleksander Stensby - NDC AI 2026 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. # 10 tips to level up your ai-assisted coding - Aleksander Stensby - NDC AI 2026 **[NDC Conferences](https://daily.dev/sources/ndc)** · 62 min read · 1 upvotes · 0 comments ## Summary A conference talk presenting practical tips for maximizing AI-assisted coding workflows. Key themes include managing context windows carefully (treating them as a finite resource), building reusable 'skills' (markdown-based slash commands) to encode preferences and workflows, always starting with a plan and persisting it, using hooks for deterministic guardrails, leveraging subagents for parallel work and infinite context, and adopting a 'compound engineering' mindset where the AI continuously learns from mistakes. The speaker also introduces concepts like loop engineering, the /insights command for session analysis, and the shift from verifying AI work to ensuring AI is doing the right work. Advice is framed around Claude Code but intended to apply to any AI coding agent. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How full should I let my AI coding agent's context window get before quality drops? Results start to deteriorate once the context window fills to around 70-80% of capacity, regardless of whether the window is the older roughly 200,000-token size or the newer million-token versions. Once that threshold is approached, harnesses auto-compact the conversation, but proactively managing context (using status line indicators, /context, or guided /compact instructions) is preferable to letting quality degrade. _Developers tuning long AI coding sessions can track context-management techniques like this on daily.dev._ ### What happens to Claude Code skill definitions when a session gets auto-compacted? Skill definitions only enter the context window when a session first starts; if the session is compacted, only the skills that were actually invoked during that session persist afterward, while unused skill definitions do not re-enter context. This trips up many users who assume all loaded skills survive compaction, so it is worth being aware of when managing long-running agent sessions. _Anyone building custom skills for coding agents can dig into nuances like this via daily.dev._ ### What is the difference between Claude Code's /goal command and just writing a detailed prompt? The /goal command lets a developer specify an outcome or objective rather than step-by-step instructions, leaving the agent to figure out how to achieve it using the tools and skills it has been equipped with. Both Codex and Claude Code implement this with a multi-agent loop where a separate agent verifies whether the goal was actually achieved, rather than letting the same agent self-report completion. _Teams shifting from prompt writing to goal-based agent workflows can follow developments like this on daily.dev._ --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#ai-coding](https://daily.dev/tags/ai-coding), [#claude-code](https://daily.dev/tags/claude-code), [#prompt-engineering](https://daily.dev/tags/prompt-engineering) [View this post on daily.dev](https://daily.dev/posts/10-tips-to-level-up-your-ai-assisted-coding---aleksander-stensby---ndc-ai-2026-kjyhgwek6) ```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":"10 tips to level up your ai-assisted coding - Aleksander Stensby - NDC AI 2026","url":"https://daily.dev/posts/10-tips-to-level-up-your-ai-assisted-coding---aleksander-stensby---ndc-ai-2026-kjyhgwek6","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/10-tips-to-level-up-your-ai-assisted-coding---aleksander-stensby---ndc-ai-2026-kjyhgwek6"},"datePublished":"2026-07-01T14:53:38.030Z","dateModified":"2026-09-14T06:47:36.971Z","description":"A conference talk presenting practical tips for maximizing AI-assisted coding workflows. Key themes include managing context windows carefully (treating them...","image":"https://i.ytimg.com/vi/K7dBRuSDWTw/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/K7dBRuSDWTw/sddefault.jpg","isAccessibleForFree":true,"articleSection":"NDC Conferences","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":"NDC Conferences","logo":"https://media.daily.dev/image/upload/s--SiIyojEj--/f_auto/v1748760369/logos/ndc","url":"https://daily.dev/sources/ndc"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/10-tips-to-level-up-your-ai-assisted-coding---aleksander-stensby---ndc-ai-2026-kjyhgwek6","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,ai-coding,claude-code,prompt-engineering","timeRequired":"PT62M","video":{"@type":"VideoObject","name":"10 tips to level up your ai-assisted coding - Aleksander Stensby - NDC AI 2026","description":"A conference talk presenting practical tips for maximizing AI-assisted coding workflows. Key themes include managing context windows carefully (treating them...","thumbnailUrl":"https://i.ytimg.com/vi/K7dBRuSDWTw/sddefault.jpg","uploadDate":"2026-07-01T14:53:38.030Z","duration":"PT62M","url":"https://api.daily.dev/r/kjYhgWEK6","embedUrl":"https://www.youtube.com/embed/K7dBRuSDWTw"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"NDC Conferences","item":"https://daily.dev/sources/ndc"},{"@type":"ListItem","position":3,"name":"10 tips to level up your ai-assisted coding - Aleksander Stensby - NDC AI 2026"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/10-tips-to-level-up-your-ai-assisted-coding---aleksander-stensby---ndc-ai-2026-kjyhgwek6#faq","mainEntity":[{"@type":"Question","name":"How full should I let my AI coding agent's context window get before quality drops?","acceptedAnswer":{"@type":"Answer","text":"Results start to deteriorate once the context window fills to around 70-80% of capacity, regardless of whether the window is the older roughly 200,000-token size or the newer million-token versions. Once that threshold is approached, harnesses auto-compact the conversation, but proactively managing context (using status line indicators, /context, or guided /compact instructions) is preferable to letting quality degrade. Developers tuning long AI coding sessions can track context-management techniques like this on daily.dev."}},{"@type":"Question","name":"What happens to Claude Code skill definitions when a session gets auto-compacted?","acceptedAnswer":{"@type":"Answer","text":"Skill definitions only enter the context window when a session first starts; if the session is compacted, only the skills that were actually invoked during that session persist afterward, while unused skill definitions do not re-enter context. This trips up many users who assume all loaded skills survive compaction, so it is worth being aware of when managing long-running agent sessions. Anyone building custom skills for coding agents can dig into nuances like this via daily.dev."}},{"@type":"Question","name":"What is the difference between Claude Code's /goal command and just writing a detailed prompt?","acceptedAnswer":{"@type":"Answer","text":"The /goal command lets a developer specify an outcome or objective rather than step-by-step instructions, leaving the agent to figure out how to achieve it using the tools and skills it has been equipped with. Both Codex and Claude Code implement this with a multi-agent loop where a separate agent verifies whether the goal was actually achieved, rather than letting the same agent self-report completion. Teams shifting from prompt writing to goal-based agent workflows can follow developments like this on daily.dev."}}]} ```