With the purchase of this course, you also get the exclusive opportunity to subscribe to the new DS live assistant app, the new app that can help data scientists during meetings as well as provide feedback, advice, etc. post-meeting. Below there is an example of a possible application using a realistic job interview from Youtube.

I picked a job interview just because it is easy to find them online, but in any case, the app is designed to work for any kind of meeting (cross-functional with many people, 1:1 with your PM or boss, etc.). That is, it will work as long as it involves DS-related conversations, such as technical deep dives, new product kick-offs, feature prioritization, etc. And at the end of the day, this mock interview is simply a deep-dive about how to kick-off an anti-spam project. A PM - DS corporate meeting on the same topic would be basically the same.

In general, I also think that watching the video below as well as reading the app interview feedback can be really useful if you are preparing for job interviews. It can help you avoid certain common mistakes.

A great feature of the app is the ability to advise not only during the meeting, but also post-meeting, just like a staff level data scientist would do. For instance, in this case you could be asking the app for feedback on your interview performance. There are several other examples on the app landing page.

That youtube mock interview is a long video, but it is very realistic. I would suggest trying to re-listen to it in light of the app feedback above. It is very enlightening and helps notice common mistakes in job interviews.

For instance, in this case the candidate doesn't do a bad job at all from a technical standpoint. I would say that the reason that she doesn't come across as doing particularly well is mostly point 1, 2, and 5 from the feedback.

She is treating the interview as an exam (question -> answer), rather than an open ended discussion whose goal is to design a solution for a current problem. That leads to not asking for clarifying questions, not engaging with the interviewer about different strategies, or ignoring crucial steps if she felt unsure about that part.

Overall, this leaves with the feeling that, by the end of the interview, we still don't really have a clear and actionable plan to solve the spam issue. And when the interviewer has that feeling, it is unlikely to be a pass, no matter how good some of the answers were (and some answers were actually good).

So, beside the real-time help, something else that the AI did really well here was translating this vague feeling after watching the interview (-> she didn't quite come up with a solution despite clearly having knowledge) into exactly where things went wrong. And this could be really useful for someone who is good technically as a data scientist, but could improve specific interview-related skills.

Matter of fact, it's often the best people who have this issue: the desire to show how good they are might lead to trying to overdo (-> I know the answer, here it is, I don't need any help), instead of working together towards a solution, asking for clarification, checking-in during the response to make sure you are going the right direction, etc.