Web scraping with AI, without paying the AI on every run
Write what you want, such as every product of a category with its name and price. AI explores the site and hands you ordinary Python you can read, edit and publish.
AI once, a program forever
The result is a Python program with Playwright. Every change makes a new numbered version, and runs of the published version use no model at all.
Learn more →Schedule it, call it, chain it
A published scraper runs on a schedule, behind an API endpoint, as a flow step or from a script with one line, and returns its rows.
Learn more →from waiso import scraper
def main(category: str = "2"):
res = scraper.run("shop_products", {"category": category}, timeout=120)
if res.not_found:
return []
return res.rows
Real sites, handled
A search with no result ends as “Nothing found”, not an error. Sites that block servers go through your proxies, and captchas through your solver.
Learn more →Questions
Does a scheduled run call the AI?
No. It runs the published program, with no model.
Can I edit the scraper’s code?
Yes. The program is plain Python; edit it by hand or ask the assistant, and each change becomes a new version.
Does Waiso provide proxies or solve captchas?
No. You bring the proxies and the captcha solver; a checkbox captcha is cleared with a click.