Real-Time Web Scraping with AI Agents: No Puppeteer, No XPath, Just a Goal

Traditional scraping breaks every time a site updates its HTML. I’ll show a different approach: give an AI agent a URL and a plain-English goal, and it figures out the page on its own.
Live on stage: a hotel search app that sends Booking.com and Agoda URLs to Tinyfish agents. Each agent autonomously navigates the real site, handles dynamic content, and returns structured JSON — no CSS selectors, no brittle DOM parsing.
II’ll also show the Tinyfish run dashboard live so you can watch the agent clicking through pages in real time.