// a hands-on course in 19 modules
From first crawl to production pipeline.
Learn Crawl4AI the way you'll actually use it: one runnable script at a time. Each module pairs a concept with real code, a video walkthrough, and a lab with explicit success criteria.
~/course — python first_crawl.py
4levels
19modules
19labs with criteria
v0.9.xAPI coverage
~/course/levels
The pipeline, in four stages
01
Beginner
First Crawl to Clean Output
Install Crawl4AI, run a single crawl, distinguish browser-level from run-level configuration, read a CrawlResult, and control caching and basic errors.
enter level 01 →
02
Intermediate
Structured Extraction & Batch Crawling
Extract structured JSON with CSS/XPath and regex, filter markdown, chunk content, and run polite concurrent batch crawls.
enter level 02 →
03
Advanced
Authentication, Anti-Bot & Rich Media
Maintain sessions, paginate, route through proxies, escalate past anti-bot systems, capture virtual-scroll feeds, parse PDFs, and inspect network traffic.
enter level 03 →
04
Expert
AI-Assisted, Observable Production Pipelines
Choose the right extraction strategy under cost constraints, build hybrid AI pipelines, run adaptive/deep crawls, observe pipelines, and extend Crawl4AI with custom strategies.
enter level 04 →
~/course/method
How each module runs
READ
Concept first. Why the API exists and where it fits, in plain engineer-to-engineer terms.
WATCH
Video at the point of need, not dumped at the end of the page.
RUN
Copy the code and run it locally. Every example is a complete script, not a fragment.
BUILD
Practice lab with success criteria. Your checklist state is saved as you go.
CHECK
Common mistakes, up front. The errors everyone hits, so you can skip them.