First stage involved a take-home machine learning assignment on a 12-way text classification problem. A potential solution is to use character-level CNN for text classification. SAP only invites you if you receive more than 70% on the test set. I got to know this by asking one of the interviewer in the second stage.
Second stage involves two consecutive rounds of in-person interviews at Singapore SAP office. Each round was 1 hour and had two interviewers. In the first round, each interviewer asked one whiteboard programming question. Questions were on merging two sorted lists and how to query a large database which keeps growing. The interviewers asked more questions on NLP as that was what they were working on at SAP. The interviewers don’t care what type of machine learning projects you have worked on. They simply asked questions pertaining to their own projects at SAP.
In the second round, I was interviewed by two different interviewers. One interviewer asked a whiteboard question on masking in Tensorflow API, whereas the other did not ask any whiteboard questions. One of the interviewer clearly informed me that they are looking for a wholesome candidate and they would like to avoid “silo candidates” who are only expert in one area such as machine learning. He wanted someone who is knowledgeable in the full software development cycle across the entire full stack.
In summary, the interviewers were totally not prepared for the interview. They did not even read my resume before the interview. I had to re-introduce myself to each of the two panels. Most of the questions were asked on the fly without any preparation. Truthfully, the interviewers are actually looking for someone who is merely a good full-stack software engineer, as can be seen by the question on querying database and the "non-silo candidate" comment. Although this job is advertised as machine learning role, you are highly likely to get this job if you are a good full-stack software engineer but with very little actual machine learning knowledge.
