For Java code, we can see at exactly which step something failed, and we catch issues a lot earlierIn terms of cross-platform testing, I have used it mostly for the web and it's pretty strong. It's probably an eight out of 10. The reason it's an eight is that we've had a couple of issues lately with videos not loading or browsers dying after some execution, although that happens very rarely. For mobile, from what I have seen, it's probably also an eight out of 10. One thing they need to have is physical devices. And I am hoping in the future that there will be more virtual devices. Also, there is some error tagging that you can do in Perfecto, but it didn't work very well when I was using it. I wish that you could tag it with an error message per test case and it would be done automatically, instead of manually. In addition, if Perfecto could use AI to say, "Your environment is stable," or "not stable," and give a score on that, that would be really helpful. It could notify us through email. That would help, aside from the failed and passed test cases. It would be something more intelligent. If there were certain tags with certain errors and there were a certain number of them, it could tell us if login was broken in the environment, for example. There's definitely a little bit of room for improvement in terms of AI.
