# Learn AI > A hands-on tour of modern AI assistants and agents — chat, coding CLI, cloud agents, UI generators, self-hosted agents. End with a tool you built. ## Curriculum ### Start Here What modern AI assistants are, and what you can build with them. - [What AI assistants are, in 3 minutes](https://learnai.databookman.com/lessons/welcome): The shortest possible intro. No jargon. (3 min) - [What can you actually build?](https://learnai.databookman.com/lessons/capabilities): A capability map — from a doc to a deployed app. (4 min) ### Foundations What every AI tutorial assumes you know — prompt, CLI, git, safety. - [Writing prompts that work](https://learnai.databookman.com/lessons/concept-prompt): The shape of a good prompt. Iterate; don't restart. The one skill that compounds. (5 min) - [The CLI / terminal](https://learnai.databookman.com/lessons/concept-cli): A text window where you tell your computer what to do. Where most AI tools live. (4 min) - [Git — your AI safety net](https://learnai.databookman.com/lessons/concept-git): Version control as 'undo for your whole project'. Essential when AI edits files. (4 min) - [Hallucinations, leaks, and what AI can't do](https://learnai.databookman.com/lessons/concept-safety): The five failure modes every AI user should know — and how to defend against them. (5 min) ### Chat assistants ChatGPT, Claude.ai, Gemini, Perplexity, Mistral Le Chat. - [Drive a chat assistant](https://learnai.databookman.com/lessons/chat-first): ChatGPT, Claude.ai, Gemini all share this shape. Try the imitation. (3 min) - [Projects, GPTs, Gems — custom assistants](https://learnai.databookman.com/lessons/chat-projects): Drop in files + instructions to make a focused helper for one job. (3 min) - [Artifacts and canvases](https://learnai.databookman.com/lessons/chat-artifacts): When the assistant produces a thing it pops into a side panel. (3 min) ### Coding assistants Cursor, Claude Code, Codex CLI, Antigravity, Windsurf, GitHub Copilot. - [What coding assistants do](https://learnai.databookman.com/lessons/code-what): An AI pair programmer that reads, edits, and runs your code. (4 min) - [Install and first run](https://learnai.databookman.com/lessons/code-install): Most CLIs follow the same install pattern. (4 min) - [Slash commands, agents, skills](https://learnai.databookman.com/lessons/code-slash): Drive a pretend coding-CLI session. (4 min) - [Skills — recipes your assistant loads on demand](https://learnai.databookman.com/lessons/code-skills): A folder of markdown that teaches your CLI a procedure. Auto-triggered by description, model-loaded as needed. (5 min) ### Visual UI generators v0.dev, Lovable, Bolt, Claude Design, Galileo. - [Sentence in, screen out](https://learnai.databookman.com/lessons/design-first): Describe a UI in plain English, see it rendered. (3 min) - [Iterating: variants and refinement](https://learnai.databookman.com/lessons/design-iterate): Swap styles, ask for a tweak, copy the code. (3 min) ### Going deeper MCP, cost, multimodal, RAG, evals — theory you'll want once you've used the tools. - [MCP — how AI plugs into your tools](https://learnai.databookman.com/lessons/concept-mcp): The standard that lets agents talk to your filesystem, Slack, GitHub, Postgres, etc. (4 min) - [Tokens, models, and what it actually costs](https://learnai.databookman.com/lessons/concept-cost): How AI is priced, where your money goes, and how to cap your spend before it surprises you. (5 min) - [Multimodal — images, voice, screen](https://learnai.databookman.com/lessons/concept-multimodal): Modern assistants take more than text. What to feed them, what to ask for, and which tool for which medium. (5 min) - [RAG — giving AI access to your data](https://learnai.databookman.com/lessons/concept-rag): How retrieval-augmented generation works, when to use it, and the simpler alternatives most people overlook. (6 min) - [Evals — knowing your AI actually works](https://learnai.databookman.com/lessons/concept-evals): How to test AI output beyond 'looks fine to me'. The boring practice that separates hobby from production. (5 min) ### Cloud agents (parallel) Devin, Claude Cowork, Copilot Workspace, Codex Cloud. - [Why run things in parallel?](https://learnai.databookman.com/lessons/cowork-what): Hand off small, well-scoped jobs to cloud agents. (3 min) - [Sending a task, reviewing the diff](https://learnai.databookman.com/lessons/cowork-flow): From prompt to pull request without a terminal. (4 min) ### Open-source autonomous agents Hermes, OpenClaw, nanoclaw, aider, OpenHands — agents you self-host. - [The open-source agent ecosystem](https://learnai.databookman.com/lessons/agents-landscape): Hermes, OpenClaw, nanoclaw, aider, OpenHands, Goose, Continue, Cline. (4 min) - [LLM APIs in 4 minutes](https://learnai.databookman.com/lessons/agents-api): OpenAI, Anthropic, Google — same shape, three vendors. (4 min) - [Set up an autonomous agent (worked example)](https://learnai.databookman.com/lessons/agents-setup): Hermes Agent install — pattern transfers to most others. (6 min) ### Ship Something Pick one. Build it. Keep it. - [Pick what to build](https://learnai.databookman.com/lessons/ship-it): Five guided paths. Pick one and follow it end-to-end in the real tool. (2 min) - [Build · A writing assistant](https://learnai.databookman.com/lessons/build-writing): A saved Project / GPT / Gem you'll use forever. ~10 min, no install. (10 min) - [Build · A live webpage](https://learnai.databookman.com/lessons/build-webpage): A working one-page site you can share. ~15 min, no install. (15 min) - [Build · A personal script](https://learnai.databookman.com/lessons/build-script): A Python tool on your machine via a coding CLI. ~20 min. (20 min) - [Build · A Slack channel digest bot](https://learnai.databookman.com/lessons/build-slackbot): A bot that posts a daily summary to a Slack channel. Real API tokens, real scheduling. ~25 min. (25 min) - [Build · A scheduled watcher](https://learnai.databookman.com/lessons/build-watcher): A Claude Cowork agent that scans a feed on a schedule and pings you when something matters. ~30 min. (30 min) ## Reference - [Cookbook](https://learnai.databookman.com/cookbook): Copy-paste starter prompts. - [Glossary](https://learnai.databookman.com/glossary): Definitions for the AI terms tutorials drop without warning. - [What's next](https://learnai.databookman.com/whats-next): Where to go after you've finished the tutorial.