From now on ScrapingAnt is AI-capable startup!
Today's release allowed us and our users to avoid the most annoying web scraping part - writing parsers.
I've done a lot of different parsers and doing this sort of work is not that hard, but you have to maintain it to make it work in production. Imagine maintaining parsers for 100 e-commerce websites.
I've been searching for the technology we used for the last 2 years. I've started from semantic parsing and up to screenshots processing, but the results were not that good to move forward with.
No more parsers!
That's a great day! 🤖