--- title: Andrej Karpathy: "99% Use AI & LLMs Wrong" (Here’s How... description: A transcript-style explainer breaks down how large language models actually work and how to use them more effectively, using ChatGPT, Claude, Gemini, Grok, and... canonical: https://daily.dev/posts/6v5qmyq9d twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Andrej Karpathy: "99% Use AI & LLMs Wrong" (Here’s How He Does It) | daily.dev og:description: A transcript-style explainer breaks down how large language models actually work and how to use them more effectively, using ChatGPT, Claude, Gemini, Grok, and... og:url: https://daily.dev/posts/6v5qmyq9d og:image: https://api.daily.dev/og/posts/6v5qmyq9d.png og:image:alt: Post cover image 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. # Andrej Karpathy: "99% Use AI & LLMs Wrong" (Here’s How He Does It) **[Showcase](https://daily.dev/sources/showcase)** · [@webkmsyed](https://daily.dev/webkmsyed) · 1 upvotes · 0 comments ## Summary A transcript-style explainer breaks down how large language models actually work and how to use them more effectively, using ChatGPT, Claude, Gemini, Grok, and Perplexity as examples. It covers the mental model of the context window as working memory, the distinction between pre-training (knowledge, with a cutoff date) and post-training (assistant persona), when to trust model knowledge versus verify it, model tiers and pricing across providers, 'thinking' models tuned via reinforcement learning for harder math/code problems, internet search tool integration, the newer 'deep research' feature that combines search and extended thinking, uploading documents/PDFs for context, using LLMs to read books and papers, and finally introduces the Python interpreter tool for tasks models can't reliably do via next-token prediction alone (like large multiplications). ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [How LLMs Work — A Visual Deep Dive](https://daily.dev/posts/how-llms-work-a-visual-deep-dive-0t0on9slu) · Hacker News · 1 upvotes · 0 comments - [How Large Language Models \(LLMs\) Work](https://daily.dev/posts/how-large-language-models-llms-work-u6fksbwh0) · C\# Corner · 0 upvotes · 0 comments - [AI models get better at math but still get low marks](https://daily.dev/posts/ai-models-get-better-at-math-but-still-get-low-marks-bugdv2bcl) · The Register · 1 upvotes · 0 comments - [The cost of thinking](https://daily.dev/posts/the-cost-of-thinking-zi8tusdea) · MIT News · 0 upvotes · 0 comments - [You’re Using ChatGPT Wrong. Here’s What Actually Works.](https://daily.dev/posts/you-re-using-chatgpt-wrong-here-s-what-actually-works--p1jd7g7wu) · Faun · 4 upvotes · 1 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#chatgpt](https://daily.dev/tags/chatgpt), [#claude](https://daily.dev/tags/claude), [#prompt-engineering](https://daily.dev/tags/prompt-engineering) [View this post on daily.dev](https://daily.dev/posts/6v5qmyq9d) ```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":"DiscussionForumPosting","mainEntityOfPage":"https://daily.dev/posts/6v5qmyq9d","headline":"Andrej Karpathy: \"99% Use AI \u0026 LLMs Wrong\" (Here’s How He Does It)","text":"Shared: Andrej Karpathy: \"99% Use AI \u0026 LLMs Wrong\" (Here’s How He Does It)","url":"https://daily.dev/posts/6v5qmyq9d","datePublished":"2026-10-27T03:19:53.280Z","dateModified":"2026-10-27T03:19:53.287Z","author":{"@type":"Person","name":"Khishamuddin Syed","url":"https://daily.dev/webkmsyed","image":"https://media.daily.dev/image/upload/s--xXLX9Wi9--/f_auto/v1777551874/avatars/avatar_VqsDuiiXz6AnTS1eKLMWz?_a=BAMAMiWQ0","description":"Frontend Design Engineer · Building @JargonIsEasy · Learn. Think. Design. Develop.","worksFor":{"@type":"Organization","name":"Jargoniseasy","logo":"https://media.daily.dev/image/upload/s--Tyfzq-Xt--/f_auto/v1763570494/companies/jargoniseasy"},"interactionStatistic":{"@type":"InteractionCounter","interactionType":{"@type":"EndorseAction"},"userInteractionCount":8060}},"image":"https://media.daily.dev/image/upload/s--ZrL_HSsR--/f_auto/v1722860399/public/Placeholder%2006","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"sharedContent":{"@type":"WebPage","url":"https://api.daily.dev/r/2oQyqS4sK"},"isPartOf":{"@type":"WebPage","url":"https://daily.dev/squads/showcase","name":"Showcase"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Showcase","item":"https://daily.dev/squads/showcase"},{"@type":"ListItem","position":3,"name":"Andrej Karpathy: \"99% Use AI \u0026 LLMs Wrong\" (Here’s How He Does It)"}]} ```