Vibe coding a spec driven web scraper, LIVE!

Can a coding agent really build a production web scraper from a single prompt? John Rooney (Developer Engagement Manager at Zyte) breaks down agentic web scraping and runs a live demo: prompt to working spider, then shows exactly where "vanilla" agents fall apart once real anti-bot, scale and data quality enter the picture. This is talk 1 from Zyte's first Developer Community Meetup. What you'll learn: What "agentic" web scraping actually means, and what it doesn't A live demo: from a single prompt to a working Scrapy spider Why coding agents can draft a scraper but struggle to ship one (anti-bot, scale, tests, low-trust output) How to keep autonomous scraping agents on the rails Chapters 0:00 Intro: what agentic web scraping means 1:30 Live demo: from a prompt to a working spider 23:00 Audience Q&A Try it yourself Zyte Web Data for Claude Code: https://www.zyte.com/agentic-web-data/ Zyte API free trial: https://www.zyte.com/zyte-api/ More from this meetup Slides: https://zytelabs.github.io/community-... Fireside chat on AI and agents: [link to Video 2] Join 2,000+ engineers in the Zyte community Community + newsletter: https://www.zyte.com/join-community/ Reddit:   / zyte   #WebScraping #AIAgents #AgenticAI #Scrapy #DataExtraction