Interview process was an absolute joke. Interviewer was already discriminating against me because I was a new grad.

First thing she told me despite already passing the technical screen was if I even felt comfortable doing the interview since I was a new grad and this interview is going to be very mid-level focused, and when I would go on to get the question right she would make up illogical statements like "this is not the expectation we are looking for as 'mid-level'" or "I wanted you to answer it using an example from your experience," or would just throw rapid-fire questions at me without even letting me finish my answer to the last question.

Explain how you would reduce latency for your model and why you would use an int quantization over something else.