I wonder how challenging it is to develop software that understands a user interface and performs actions, even with significant changes to the interface while retaining its concept?
The software can move the cursor, click, type text, and take screenshots for analysis. It can build a knowledge base and use rules created by humans. Ultimately, the software should be able to place an order in an online store that it has never seen before, but has seen similar ones.
A terribly interesting task.
Hi, I'm Rauf Aliev. 