We keep working on web scraping API with custom-made mobile proxy pool: https://scrapingfish.com/

There is no AI in it so far but we consider adding support for parsing the result to extract data using LLM.