--- title: Hybrid and Local AI course at DeepLearning.AI | daily.dev description: JetBrains partnered with DeepLearning.AI on a free course, AI Coding Workflows: Hybrid to Local, taught using PyCharm and its AI Chat. The course starts with... canonical: https://daily.dev/posts/hybrid-and-local-ai-course-at-deeplearning-ai-z0agdjjhu twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Hybrid and Local AI course at DeepLearning.AI | daily.dev og:description: JetBrains partnered with DeepLearning.AI on a free course, AI Coding Workflows: Hybrid to Local, taught using PyCharm and its AI Chat. The course starts with... og:url: https://daily.dev/posts/hybrid-and-local-ai-course-at-deeplearning-ai-z0agdjjhu og:image: https://api.daily.dev/og/posts/Z0aGdjJhU.png og:image:alt: Hybrid and Local AI course at DeepLearning.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. # Hybrid and Local AI course at DeepLearning.AI **[JetBrains](https://daily.dev/sources/jetbrains)** · 4 min read · 1 upvotes · 0 comments ## Summary JetBrains partnered with DeepLearning.AI on a free course, AI Coding Workflows: Hybrid to Local, taught using PyCharm and its AI Chat. The course starts with Claude Code and subagents to teach spec-driven development, cost reduction, and context isolation, using Opus as a 'big brain' and Haiku as a 'little brain' implementer. It then moves to OpenCode with OpenRouter for inference routing and DeepSeek as a cheaper model, before progressing to fully local setups using LM Studio, Gemma 4 12B, DeepSeek v4 Flash, and Qwen 3.5 27B on a 32 GB laptop. Each lesson tracks metrics like tokens, turns, cost, and wall time to demonstrate how smaller and local models perform with proper guardrails and task decomposition. ## 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 reduce Claude Code costs by using subagents with different models? Use a larger, more capable model like Opus for the main conversation and planning, then delegate implementation work to a fresh subagent running a cheaper model like Haiku. This context isolation keeps token usage down while letting the bigger model handle upfront spec work and right-sized steps for the smaller model to execute. _developers optimizing AI coding costs can track new subagent and model strategies through daily.dev._ ### What local models can run AI coding agents on a 32 GB laptop? Gemma 4 12B running on LM Studio as the inference server works as a starting local model for a 32 GB laptop setup, and Qwen 3.5 27B has also been tested as a fully local 'big brain' model with good results when combined with guardrails like spec-driven development and subagents. _daily.dev helps developers exploring local-first AI coding setups keep up with new model options._ ### What is spec-driven development and why does it help smaller AI models code better? Spec-driven development divides open-ended coding tasks into smaller, bounded steps, which keeps smaller or cheaper AI models on track since they struggle with big, open-ended vibe coding. A larger model can do the upfront planning work to form right-sized steps that a smaller implementer model then executes reliably. _daily.dev keeps developers weighing agent workflows and model tradeoffs informed on spec-driven techniques._ ## Similar posts on daily.dev - [I used only local AI for a week and learned what actually needs the cloud](https://daily.dev/posts/i-used-only-local-ai-for-a-week-and-learned-what-actually-needs-the-cloud-z17uaq7bb) · XDA Developers · 0 upvotes · 0 comments --- Tags: [#devtools](https://daily.dev/tags/devtools), [#claude-code](https://daily.dev/tags/claude-code), [#local-ai](https://daily.dev/tags/local-ai), [#ai-inference](https://daily.dev/tags/ai-inference), [#opencode](https://daily.dev/tags/opencode) [View this post on daily.dev](https://daily.dev/posts/hybrid-and-local-ai-course-at-deeplearning-ai-z0agdjjhu) ```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":"Hybrid and Local AI course at DeepLearning.AI","url":"https://daily.dev/posts/hybrid-and-local-ai-course-at-deeplearning-ai-z0agdjjhu","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/hybrid-and-local-ai-course-at-deeplearning-ai-z0agdjjhu"},"datePublished":"2026-08-13T11:31:02.568Z","dateModified":"2026-09-14T09:32:18.552Z","description":"JetBrains partnered with DeepLearning.AI on a free course, AI Coding Workflows: Hybrid to Local, taught using PyCharm and its AI Chat. The course starts with...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/23b53c7a5fe7415b4cb55121ae7ae6ba?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/23b53c7a5fe7415b4cb55121ae7ae6ba?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"JetBrains","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":"JetBrains","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/53ecf0c676f34b3896ee109609d91efa","url":"https://daily.dev/sources/jetbrains"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/hybrid-and-local-ai-course-at-deeplearning-ai-z0agdjjhu","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"devtools,claude-code,local-ai,ai-inference,opencode","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"JetBrains","item":"https://daily.dev/sources/jetbrains"},{"@type":"ListItem","position":3,"name":"Hybrid and Local AI course at DeepLearning.AI"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/hybrid-and-local-ai-course-at-deeplearning-ai-z0agdjjhu#faq","mainEntity":[{"@type":"Question","name":"How can I reduce Claude Code costs by using subagents with different models?","acceptedAnswer":{"@type":"Answer","text":"Use a larger, more capable model like Opus for the main conversation and planning, then delegate implementation work to a fresh subagent running a cheaper model like Haiku. This context isolation keeps token usage down while letting the bigger model handle upfront spec work and right-sized steps for the smaller model to execute. developers optimizing AI coding costs can track new subagent and model strategies through daily.dev."}},{"@type":"Question","name":"What local models can run AI coding agents on a 32 GB laptop?","acceptedAnswer":{"@type":"Answer","text":"Gemma 4 12B running on LM Studio as the inference server works as a starting local model for a 32 GB laptop setup, and Qwen 3.5 27B has also been tested as a fully local 'big brain' model with good results when combined with guardrails like spec-driven development and subagents. daily.dev helps developers exploring local-first AI coding setups keep up with new model options."}},{"@type":"Question","name":"What is spec-driven development and why does it help smaller AI models code better?","acceptedAnswer":{"@type":"Answer","text":"Spec-driven development divides open-ended coding tasks into smaller, bounded steps, which keeps smaller or cheaper AI models on track since they struggle with big, open-ended vibe coding. A larger model can do the upfront planning work to form right-sized steps that a smaller implementer model then executes reliably. daily.dev keeps developers weighing agent workflows and model tradeoffs informed on spec-driven techniques."}}]} ```