Qualcomm interview questions

Implement Causal Masking in decoder only architecture using Pytorch. What is grouped query attention. What is Mixture of Experts. Difference between encoder only architecture and decoder only architecture. Explain convolution operation in CNNs. What is NMS algorithm in YOLO

Had an online call with the hiring manager. 1 hour technical interview, goes like this; my technical past first, then some technical questions, then some questions I had. There can be more rounds. Overall positive experience, kind and professional.