There were three rounds after the recruiter screen.
First was a coding/design round. Second was a hiring manager round. Third was a design + LeetCode-style round.
In the third round, there were two interviewers. I believe one was shadowing. The main interviewer came late.
We started with an ML design question. This was so out of touch from ML, I don't even know how he was senior ML at Comcast. Anyway, he kept combating and interrupting me the whole time. At one point, he suggested using information gain as a loss function in a neural network.
Anyway, then there was a LeetCode-style question in the second half of the round, for which he didn't even consider giving two minutes extra, even though he joined late.
Print matrix in spiral form.
The following metrics were computed from 1 interview experience for the Comcast Machine Learning Engineer role in Washington, District of Columbia.
Comcast's interview process for their Machine Learning Engineer roles in Washington, District of Columbia is extremely selective, failing the vast majority of engineers.
Candidates reported having very negative feelings for Comcast's Machine Learning Engineer interview process in Washington, District of Columbia.