Generative AI & LLMsDeepSeek R1 is an open-weight reasoning-focused large language model released by China's DeepSeek in January 2025. It achieves OpenAI o1-level performance using a novel reinforcement-learning-only training approach and is distributed under the MIT License.
Development Methods & Design PatternsA slash command is a chat shortcut triggered by typing '/' followed by a command name. Widely used in Claude Code, Discord, and Slack, slash commands let users invoke predefined prompts or actions instantly. Learn the concept, creation steps, and real-world usage.
Generative AI & LLMsA subagent is a specialized AI agent spawned by Claude Code that runs in its own isolated context. Learn how subagents enable parallel processing, specialized roles, and modular automation for complex engineering tasks, based on Anthropic's official specification.
AI & Machine LearningClaude in Excel is Anthropic's Excel add-in that brings Claude directly into the sidebar of Microsoft Excel. It reads multi-tab workbooks, explains calculations with cell-level citations, and updates assumptions while preserving formula dependencies. This guide covers pricing, Copilot comparison, and more.
AI & Machine LearningClaude in Chrome is Anthropic's browser extension that puts an AI agent directly inside Google Chrome. It navigates, clicks, fills forms, extracts data, and runs multi-step web workflows from natural language instructions.
Machine Learning & Deep LearningLoRA (Low-Rank Adaptation) is a parameter-efficient fine-tuning technique that keeps a pre-trained model frozen while training small low-rank matrices. This guide covers pronunciation, the rank decomposition math, step-by-step usage with Hugging Face PEFT, QLoRA variants, comparison with full fine-tuning, memory savings, and common implementation pitfalls.
Generative AI & LLMsLlama 4 is Meta's open-weight family introduced in April 2025, spanning Scout (17B active, 10M context), Maverick (17B active, 400B total), and the unreleased Behemoth (~2T total). This guide covers pronunciation, the Mixture-of-Experts architecture, multimodal input, licensing, deployment options, Muse Spark's strategic pivot, and how Llama 4 compares to Claude and GPT.
Generative AI & LLMsGemini 2.5 is Google's thinking-family AI model, reasoning through its own thoughts before responding. This guide covers the Pro, Flash, and Flash-Lite tiers, architecture, multimodal inputs, long-context handling, pricing, Gemini 3 successor, and direct comparisons with Claude and GPT models.
Generative AI & LLMsAgent Skills are Anthropic's open standard for packaging instructions, scripts, and reference data into folders that Claude can discover and load on demand. This guide covers pronunciation, the SKILL.md manifest, how skills differ from tools and MCP servers, authoring examples, common misconceptions, and the portability implications of the open skill spec.
Generative AI & LLMsClaude Desktop is Anthropic's official native client for macOS and Windows, bundling Claude chat, Cowork agent mode, Claude Code, MCP connectors, and local file access into a single application. This guide covers pronunciation, installation, pricing, key features, common misconceptions, and how Claude Desktop compares to the web version.
