Hi folks 🙂 I am going to introduce some useful resources for preparing data science interviews or just basically learning data analytics. You don’t need to follow all of them. Feel free to start with the materials or courses that you feel that is most suitable for your learning style or current circumstances.
Let’s begin!
Mathematics, Statistics, and Probability
1. Math:
- Mathematics for Machine Learning on Coursera:
During the job searching, you might want to review some basic calculus and linear algebra. I would recommend this course. There are three courses in this specialization, and I reviewed the Linear Algebra one while I was searching. The content is tailored to the application of math on data science or machine learning, so you will only learn essential knowledge. It’s pretty useful and practical for data analyst, scientist, or engineers to review the related concepts. However, if you want a more systematic and detailed learning experience in Linear Algebra or Multivariate Calculus, you should consult other courses.
2. Statistics:
This book is amazing! I think I might review it for maybe more than three times. I printed it out and brought it with me. I will recommend read through these chapters:
C2. Statistical Learning, C3. Linear Regression, C4 Classification, C5 Resampling Models, C6 Regularization, C8 Tree-based, C9 SVM, and C10 Unsupervised Learning
It also provides R lab so it will be good time to practice modeling in R.
It is a Youtube channel of Pro. Starmer at UNC-Chapel Hill. He is super funny! He systematically introduce basic ideas and concepts in Stats, Data Science, and ML. You could learn extremely important topics in just 5-10 minutes with cute (or dorky lol) visualizations. He explains complicated concepts in a succinct way that I regret not knowing his channel when I struggled with statistics in undergrad.
3. A/B Testing
Such a great course about A/B testing! I learn a lot from these intelligent lecturers. A must for understanding A/B testing. It might be a little bit difficult to follow their conversation though, so make sure you take notes.
4. Probability
It is a great website for learning math, stats, and probability. I did some brain-teaser related to probability and Bayes theory. The UI is well-designed and easy to use. They also have their own Wiki so you can learn with clear explanations about concepts or questions. Its premium content is not free.
Machine learning
- Pattern Recognition and Machine Learning, Bishop C.M.
I first knew this book from the ML course I took from my last semester during master’s. It is well-written and very famous within the data science community. I would recommend to read with ISLR.
Programming Skills
- SQL
I would recommend to use SQL ZOO, SQL introduction at Mode Analytics, Hackerank, and Leetcode to practice SQL. It is ESSENTIAL to have SQL skills. I would do 5~7 SQL quizzes every day just so I am keep up my familiarity with it since I don’t use it often at my current job. Make sure you know join, subquery, and window functions.
- R
I didn’t practice R too much because I use it a lot. If you want to learn R, feel free to message me. I hosted R workshops for my interns last summer. I’d love to share.
- Python
Since Python is my second language (my first is R), it took me sometime to practice Python. I just switched my work from R to Python so I could have chances practicing it. You can also find some interesting dataset and play around using Python. Also, StackOverflow and Google are your good friends! Just searching the solution when you are stuck.
Besides, I tried to practice some algorithm questions in python on Leetcode, but I would say it was extremely difficult for me since I didn’t know much about algorithms. I might not be able to provide advice regarding this matter.
Business and Product Sense
- Cracking the PM Interview
It is useful to learn more about product and business. It has chapters related to how to practice interviews for consultants, which is not exactly the way we do in the data science world, but I would say it would be helpful when being asked some project workflow questions. Like you have to understand the business problem first, then break down into smaller analytical components.
- Data Science for Business
Well-written book about how DS is applied to understand business and provide recommendations. It definitely inspires me to try more different methods.
This is a series of multiple articles related to product and data, written by the ds team at Sequoia. TREASURE, period. They introduce important ideas around product, including how to measure the success of your product, to increase product growth, and to improve engagement, etc. Once you finish the whole series, you will have a very clear picture about the product workflow, and what role data scientists play at this process.
Others
- Resume is the key to get the interview. Make sure you have people editing your resume.
- Practice behavioral questions. Your personal pitch should be around 30 secs (Questions like tell me about yourself or introduce yourself). You can create an excel chart with your project experiences, leadership, teamwork, success, etc.
- Mock interviews are helpful! Practice behavioral and technical interview questions. Try them on whiteboard to be familiarize with coding tests.
- Keep track of your applications so that you will have a clear picture of your timeline.
- Make sure you polish your Linkedin profile. For networking, you can just message people on Linkedin and tell them you are interested in grabbing a cup of coffee. I wrote a post about networking a year ago, feel free to check it out (it’s written in Mandarin though). Also, polish your Github profile, too.
Thank you for reading! Feel free to ask questions or leave any comment. Have a good day and stay healthy.
