Karovia/fullstack-ai-agent-roadmap

What it solves

Learning full-stack development plus AI Agents is overwhelming and scattered. This project replaces scattered link lists with a single, structured, 18-month roadmap that takes a complete beginner from zero programming knowledge to being able to independently design, build, and ship AI Agent products.

How it works

It's a collection of over 110 Markdown tutorials (58万字, about 580,000 Chinese characters) organized in Obsidian. Each concept is explained with everyday analogies, complete runnable code with line-by-line comments, expected output, and practice exercises. The roadmap is divided into 10 chapters (00–09), each with a hands-on capstone project (for example, building a Python crawler, a mini React, a backend API, an AI document Q&A system, and finally a full product). Every chapter also includes a curated list of 400+ GitHub projects, with guidance on how to learn from each one. There are also Obsidian Canvas mind maps for a global view, plus templates for learning contracts, weekly reports, and milestones.

Who it's for

This is for absolute beginners with no prior knowledge. It's especially suited for students, self-learners, and career-switchers who want to become full-stack engineers in the AI era. The recommended pace is about 3 hours per day (or 1.5 hours if working full-time), with a total duration of 8 to 24 months depending on available time.

Highlights

• 110+ tutorials, 580,000+ characters, all in Markdown, ready for Obsidian
• Full runnable code with explanations, outputs, and practice answers
• 30+ chapter projects — you must complete each to move on
• 400+ curated GitHub projects, organized by difficulty level
• Obsidian Canvas mind maps for navigation and chapter relationships
• Built-in learning contract, weekly report, and milestone templates
• Four learning principles: no shortcuts, project-driven, Feynman technique, and build-your-own (mini React, mini Agent SDK)