Agreed. I found that its suggestions constantly break flow as you’re now evaluating something that was pushed on you. Beyond basic algorithms and boilerplate c

Agreed. I found that its suggestions constantly break flow as you’re now evaluating something that was pushed on you. Beyond basic algorithms and boilerplate code completion (which would arguably be better implemented via better intellisense than ML), it was basically zero help. Even in the areas it helped, time to eval mostly equaled the time I would have taken to write the same thing manually.