AI Skills

🔅 Applied AI: Getting Started with Hugging Face Transformers 📝 Get up and running with pretrained transformers in Hugging Face, the popular platform for natural language processing (NLP) applications. 🌐 Author: Kumaran Ponnambalam 🔰 Level: Intermediate ⏰ Duration: 1h 14m 📋 Topics: Hugging Face Products, Applied Machine Learning, Artificial Intelligence

🔰 Ultimate Web Development Course 2026 - Build Modern Websites 🌟 4.5 - 1646 votes 💰 Original Price: $19.99

📖 This is the only modern, always up-to-date Web Development course you will ever need.

🔊 Taught By: Haris Ali Khan 📤 Download Full Course 📤 Download All Courses

📊 Free collection teaching AI coding agents senior engineering skills. Named agent-skills. 60,800+ GitHub stars. Integrates with Claude Code, Codex, Cursor, or Gemini CLI. Features: → /spec: Define requirements before coding. Always. → /plan: Break specs into atomic tasks. No giant PRs or mystery diffs. → /build: Implement one slice at a time. Test-driven and individually committed. → /build auto: Generate plan and run tasks in one pass. Single approval. Autonomous execution. Pauses on failures or risky steps. → /test: Verify code works. Tests are proof, not afterthoughts. → /review: Enforce code health pre-merge. Real quality gate. → /code-simplify: Rewrite for clarity. Remove cleverness. → /ship: Run full production checklist. Speed requires no skipped steps. Context-aware activation: Building an API triggers api-and-interface-design. Building UI triggers frontend-ui-engineering. No config needed. 100% Open Source. Repo: https://github.com/rosnjs/agent-skills

💡 What Are LLMs? 📊 Large Language Models are AI systems trained on vast text data to understand and generate human-like language. 🧬 Built on transformer architecture, they predict the next word using patterns in grammar, context, and knowledge. ⚡️ From writing emails to coding and reasoning, they power tools like chatbots and assistants. 🔥 Flaws like bias exist, but they’re reshaping how machines think. Language is the new code.

SHOCKING: 99% of GTM engineers using Fable 5 are installing skills wrong. Right now, every GTM engineer on LinkedIn is screaming "skills, skills, skills"... But here's the truth: dropping a single SKILL.md file into a folder and calling it a skill is not how Anthropic builds them internally. And it is why most skills underperform. To build a skill stack that actually compounds, you need to master: - The folder architecture Anthropic uses internally: scripts, assets, reference files, and examples, not just a markdown file - The gotchas section that Anthropic calls the highest-signal content in any skill, built from real failures not upfront warnings I spent weeks sourcing, testing, and documenting the exact system and compiled every skill source, install command, chain workflow, and CLAUDE.md block into one complete bank. I'll give it to only 4,500 people. To get it: Follow me MUST (so I can DM) Comment "Agents" I'll DM you the document If you don't follow or comment, you won't receive it.

Local AI Stack in 2026: what you can actually run on a laptop for text, video, RAG and notebooks Main point: local AI is no longer a weekend toy. The useful setup is not the biggest model, but the right model for the job and hardware. 🧩 Text: start with Qwen3-4B/8B, Gemma-3-4B, or Llama-3.2-1B/3B. Qwen3 is neat because it has /think and /no_think: use slower reasoning only when needed. MiMo is worth watching too: Xiaomi's MiMo-7B-RL is on GitHub/HuggingFace, tuned for math, code and reasoning. The paper says the base model used 25T pretraining tokens, then RL on 130K verifiable math/code tasks. Video: Lightricks/LTX-Video and LTXV-13B can run locally through Python/ComfyUI, but be honest with your laptop. The 13B line wants a serious GPU. For experiments, start with distilled/FP8 or the 2B branch. Lower quality, much faster iteration. Your docs: local RAG means Chroma or LanceDB, Ollama embeddings like embeddinggemma or qwen3-embedding, then a small LLM. Important detail: use the same embedding model for indexing and search, or the answers will sound smart but miss the source. Jupyter AI also fits the stack: chat inside JupyterLab, attach files, ask about a notebook or cell, and connect it to local Ollama or vLLM. ⚠️ Hardware note: 16 GB RAM is fine for 1B to 4B quantized models. 32 GB RAM or a discrete GPU makes 7B to 8B much nicer. Long context eats memory fast: Ollama defaults to 4096 tokens, and raising num_ctx hits RAM/VRAM. Best 2026 laptop stack: small LLM, local embeddings, RAG, Jupyter or IDE integration. You can build it without cloud calls and without a token bill.

You install Claude Code and stop there. Here are 24 things (actually) worth adding: If you're totally new to Claude Code: Start here: lnkd.in/emwZS5yS Then read: lnkd.in/eG2-Fj6S Plug-Ins (bundled tools, agents, and commands) gstack — 23 specialist dev tools in one install Install → github.com/garrytan/gstack (82.7k★) superpowers — complete dev methodology, 14 skills Install → lnkd.in/eppbgRaK (192k★) codex-plugin-cc — OpenAI's official Codex plugin Install → lnkd.in/eTweEPmw (8.9k★) financial-services — IB, PE, equity, wealth Install → lnkd.in/e9fpC2XF (23k★) claude-for-legal — legal workflows, every practice area Install → lnkd.in/eYgW_QUy (6.6k★) claude-skills — 263+ skills across every platform Install → lnkd.in/eYXHrn27 (5.2k★) marketingskills — 40 marketing tools, full growth ops Install → lnkd.in/egt-7ZwM (28.8k★) social-media-skills — my content OS. Posts, reels Install → lnkd.in/emDvetxm Skills (specialist instruction Claude loads on demand) frontend-design — kills generic AI UI Install → lnkd.in/eMpNx__b (277k installs) hyperframes — write HTML, render video, agent-native Install → lnkd.in/ed-wSdsx (18.6k★) ai-second-brain — Karpathy-style wiki, AI history Install → lnkd.in/et2waZ79 notebooklm-skill — Claude queries your research Install → lnkd.in/edUnrPTe humanizer — strips AI writing tells from any draft Install → lnkd.in/eekWNVYm (2.9k★) claude-seo — GEO-first SEO skill, built for the AI era Install → lnkd.in/ec5AZ_pW (4.5k★) antfu-skills — Vue and Vite core team skills Install → github.com/antfu/skills (3.5k★) caveman — cuts 65% of tokens, talks like caveman Install → lnkd.in/e4nxpEJi (59.2k★) MCP Servers (live connections to your apps) granola — meeting notes fed to Claude Install → lnkd.in/e23ayrFh slack — reads channels, posts updates Install → lnkd.in/exz5AtNM notion — reads and writes your docs Install → lnkd.in/e6HXirqR kondo — triages your LinkedIn DMs Install → relay.trykondo.com/mcp zapier — 9,000+ apps, one connection Install → mcp.zapier.com higgsfield — cinematic video from a prompt Install → higgsfield.ai/mcp perplexity — live web search for Claude Install → lnkd.in/ejdkVnes agent-browser — browser automation, fewer tokens Install → lnkd.in/eUS4cxjs (22k★) Save this. Come back when you set up Claude Code properly. Repost ♻️ to help someone in your network. P.S. Which one are you installing first?

🔅 AI Mastery: LLMs Explained with Math (Transformers, Attention Mechanisms & More) ⏲ 5 hours 📁 34 Lessons

📔 Unlock the secrets behind transformers like GPT and BERT. Learn tokenization, attention mechanisms, positional encoding, and embedding to build and innovate with advanced AI. Excel in the field of machine learning and become a top-tier AI expert.

🎙 Taught by: Patrik Szepesi 📤 Download All Courses

🔅 AI-Powered Software Development: Coding, Testing, and System Design 📝 Explore how to effectively leverage generative AI tools across the development lifecycle—from coding and testing to architecture design and agile project management. 🌐 Author: Shaun Wassell 🔰 Level: Intermediate ⏰ Duration: 2h 46m 📋 Topics: AI Software Development, Generative AI Tools, Software Development 🔗 Join Artificial intelligence for more courses

💻CloudReve 🛠 CloudReve is a self-hosted file management system with multi-cloud support, offering a wide range of features for storing and organizing data. 🔰The system supports various storage backends, upload/download speed limiting, and integration with Aria2. 🔰Users can easily manage files via WebDAV and drag & drop, generate time-limited sharing links, and preview files of various formats online. 🔰CloudReve also allows theme customization and supports multi-user mode, making it a versatile tool for file management. 🔗Links: https://github.com/cloudreve/Cloudreve?tab=readme-ov-file

🔎We found a tool for searching PDF files for you 🛠 PDF Search is a document search engine that lets you browse over 18 million PDF documents. 🔰 She will find articles, guides, courses, scientific materials and much more. ⚙️ Some of the service's features: 🔹 Organize your search using smart tags. Click on the tag next to the document preview to find more relevant documents; 🔹 Interactive search. You can find not only a document preview and smart tags, but also a detailed summary and a selection of the most important facts within the document; 🔹 Natural language processing. The service allows you to view rich semantic metadata extracted from thousands of documents; 🔹 Find facts and entities, not just text. The tool goes beyond classic entity identification and returns facts and events hidden in the content. 🔗Links: https://www.pdfsearch.io/index.php

In recent times, the popularity of transformer-based LLMs and LLM applications such as AI agents has skyrocketed. Compute is in high demand, while models soar in parameter count—reaching hundreds of billions and trillions of parameters in the largest LLMs. Luckily, researchers have been moving towards techniques to reduce the compute and VRAM needed to store, train, and run models. This is where small language models (SLMs) come in. Small language models are neural language models that are much smaller in size (typically billions of parameters or fewer) than today’s massive LLMs (which often have hundreds of billions). By design, SLMs can run on consumer-grade devices like smartphones, embedded systems, or PCs, offering fast inference and a much lower cost. Researchers often consider models under about 10 billion parameters to be SLMs, since such models can fit on common hardware with low latency.

In June of 2025, Nvidia research released a paper detailing the potential of SLMs, titled “Small Language Models are the Future of Agentic AI.” One of the key takeaways from the paper is that since agents are typically tailored towards solving very specific tasks, a full hundred billion parameter LLM is not required to be proficient at the task. They show that SLMs are in fact enough for specific agentic applications with examples in specific industries. SLMs use many state-of-the-art optimization techniques and fine-tuning to decrease the model size and improve efficiency. Some of these techniques allow SLMs to be decently powerful and useful at small sizes. Techniques include quantization, mixture-of-experts (MoE), low rank adaptation (LoRA), pruning, flashattention and more.