Initial phone screen followed by a take home data challenge, then another interview with the data science hiring manager. This would be followed by an onsite or onsites. In general they were very nice. However, their data science interview process isn't very good. This was also noted by another interviewee. The take home data challenge is fine, but the rubric for assessing it is a bit nebulous. Furthermore, they do not address the work you do in the take home challenge in the next interview with the hiring manager. A lot of companies do this, and it is I think a waste of candidates' time and effort. If there is anything candidates can do in the data science interview process that looks anything like the work they will actually do, it is the take home data challenge. The question, or 'gotcha,' that is asked during the technical interview with the data science hiring manager is perhaps a 'knowledge' question like the other interviewee alluded to. I initially was a bit hazy on where exactly they were going with the question, but I mentioned it looked kind of like reservoir sampling. However, I only vaguely remembered reservoir sampling. The hiring manager even addressed before asking the question that it isn't really the best type of question to ask over the phone. So why do they ask it? It is definitely an irrelevant question since it is really a poor proxy to determine whether a candidate could actually perform well as a data scientist. Just a bad question. Asana is not the only company that asks stuff like this, but that is no justification, companies really need to have have better, i.e. relevant assessments of data science candidates; there is really no excuse for this. If future data scientists read this, and have had similar experiences, I think they (you) should write about your experience. Companies really need to get their act together. Ironically, of all the disciplines that have an interview process one would think that data science would have a better approach to hiring and interviewing; experimental design and counterfactuals are a big part of data science. Finally, I was sent a generic rejection email, but the hiring manager did say the initial recruiter would get back to me. If you say you are going to do something, do it.