Tips for a Data Science job interview
1. Brush up on basic maths and statistical concepts
You don´t want to stutter when asked about basic concepts such as ROC curves, logistic regression, etc. Can you explain the difference between supervised and unsupervised learning both to a child and a math major? Good!
2. Prepare a set of practical examples to illustrate your value
Data Science is all about bridging the gap between the world of tech and business. As a candidate, you are hence expected to be able to use scientific methods to solve real business problems. Sometimes even particular IT skills and tech knowledge (e. g. SQL, programming, R) are required or recommended. Prepare to talk about past research questions that you have successfully answered and to explain the hows and whys in-depth, including tools and skills you have used.
3. If you have knowledge gaps, admit them
If the interviewer asks you about a specific topic, you only have a faint grasp of, admit it. Save yourself the pain of clueless babbling about an unfamiliar subject and be honest about what you do not know. Then demonstrate enthusiasm about learning new things (eg., language, technology). You will come across as more trustworthy and reliable.
4. Think about what is your favourite industry/research area to apply Data Science and why?
Employers usually know well what they are looking for in a candidate. But equally, do you know what you want? What hopes and ideas are you bringing to the table? What is your favourite area for applying data science, and why? Will you be ok doing a lot of data engineering, or do you actually just want to work with fancier ML algorithms and neural networks? These are important questions to think about.
5. Learn to rationalize and defend your decisions
At any data science interview, you will be expected to dissect problems in detail and come up with potential solutions. The interviewer will be interested in hearing your take on the problems just as much as following your thought process. Make it easier for them! Structure your ideas, back up with logical arguments, and explain your solution choices.