Free guide · Node.js · React · Next.js
Become an AI-powered full-stack engineer
A visual, beginner-to-expert path through LLM apps, RAG, agents, MCP, evals, and production engineering, all in TypeScript with the stack you already use.
- Start Here2 weeksbeginner Understand the role and pick a study plan.
- Foundations4 weeksbeginner TypeScript, Node.js, React, databases, APIs, and DevOps.
- LLM Fundamentals2 weeksbeginner Tokens, context windows, choosing models, keys and billing.
- LLM App Engineering5 weeksintermediate Streaming, structured outputs, tool use, RAG, and caching.
- Agents & Orchestration5 weeksadvanced Agent loops, multi-agent systems, MCP, durable execution.
- Production AI2 weeksadvanced Evals, tracing, reliability, security, and cost.
- AI-Assisted Development1 weekintermediate Build, review, test, and debug with coding agents.
- Projects3 weeksadvanced Six builds, ending with a production SaaS assistant.
- Career2 weeksintermediate Portfolio, writing, and AI system design interviews.
Show text descriptionHide text description
The roadmap is an ordered path of 9 stages. Each stage builds on the ones before it. Durations follow the 6-month standard plan; the 3-month plan compresses them and the 12-month plan expands the foundations.
- Start Here (2 weeks, beginner): Understand the role and pick a study plan.
- Foundations (4 weeks, beginner): TypeScript, Node.js, React, databases, APIs, and DevOps.
- LLM Fundamentals (2 weeks, beginner): Tokens, context windows, choosing models, keys and billing.
- LLM App Engineering (5 weeks, intermediate): Streaming, structured outputs, tool use, RAG, and caching.
- Agents & Orchestration (5 weeks, advanced): Agent loops, multi-agent systems, MCP, durable execution.
- Production AI (2 weeks, advanced): Evals, tracing, reliability, security, and cost.
- AI-Assisted Development (1 week, intermediate): Build, review, test, and debug with coding agents.
- Projects (3 weeks, advanced): Six builds, ending with a production SaaS assistant.
- Career (2 weeks, intermediate): Portfolio, writing, and AI system design interviews.
Who this is for
Web developers adding AI
You ship React and Node.js apps and want to build features on top of LLMs that hold up in production, not just demos.
Career switchers and students
You are starting from the basics and want one ordered path from JavaScript fundamentals to agents, evals, and a hireable portfolio.
Engineers leveling up
You have called an LLM API before and want depth: retrieval quality, agent design, tracing, cost control, and security.
Pick a pace: 3, 6, or 12 months
Same destination, different starting points. Each plan ends with deployed projects and interview preparation.
3 months
Fast track
For: Already comfortable with TypeScript, React, and a backend.
Focus: LLM fundamentals, application engineering, one agent project, evals, and a production-grade capstone.
10–12 h/week
6 months
Standard path
For: You know JavaScript but have gaps in Node.js, databases, or deployment.
Focus: Close foundation gaps first, then work through every LLM, agent, and production section with all projects.
8–10 h/week
12 months
From zero
For: New to programming or coming from a non-web background.
Focus: A full foundations semester, then the AI sections at a steady pace, finishing with a portfolio and interview prep.
6–8 h/week
How this wiki teaches
Visual first
Every concept starts with a diagram built as code: accessible, crisp at any size, and correct in light and dark mode.
Runnable TypeScript
Modern ESM, async/await, and strict types. Every snippet runs as shown, so you learn by building, not by guessing.
Production focus
Evals, tracing, retries, prompt-injection defenses, and cost engineering are core curriculum, not an appendix.
Portfolio projects
Six step-by-step builds of increasing difficulty, ending with a SaaS AI assistant you can show in interviews.