How to prepare for Data Science Interviews

Kushagra Mittal
3 min readDec 1, 2020
Photo by Steve Halama on Unsplash

First of all, if you’re preparing for any data science interview, you must know about job roles and responsibilities. Under data science, there are many roles are offered by companies like Data Scientist, Data Analyst, Data Architect, Machine Learning Engineer, Business Analyst, Marketing Analyst and many more. These roles are depends upon your skills and experiences which you have.

Note : Data Science Interviews completely depends upon for what type of job role you are applying to company .

But apart from this, I am gonna give you some tips and topics from which most questions asked during interview.

  1. Thoroughly, read your resume because if you’re are applied for company through an LinkedIn or Glassdoor, they will shortlist you through your resume and they will definitely asked from resume about your projects, skills, Internship/Training that you had mentioned in your resume.

Note : Don’t try to input any fake skills or internships or any other things.

Be yourself what you are.

2. Fundamentals of Programming Language like python/R which you know well or used in your data science mini-projects that you had mentioned in your resume.

3. There are some basic terminologies which you must know and have understanding in that topics. Terms like ML, AI , Deep Learning and their applications in the real life or digital trend you must know.

i. Types of learning comes under ML and what algorithms comes under different learnings.

ii. Difference between the Supervised and Unsupervised and Reinforcement Learning

iii. Regression Vs Classification

iv. Basics of Linear Regression, Logistic Regression

v. Underfit Vs Overfit

vi. Regularization, how to optimize the algorithm — cost function, gradient descent , learning parameter.

Note: You can also enroll in Andrew Ng ML Course on Coursera for better understanding refer Machine Learning by Stanford University | Coursera

4. How to deal with missing values in datasets, how to choose features, remove outliers, On what basis you choose any algorithm.

5. Various visualization libraries you had used and different types of graphs.

6. If you had mentioned or done any projects in NLP then,

i. What is NLP ? What are application of NLP in real life ?

ii. Some basic terminology which you must know like Stemming, Tokenization, Lemmatization, Vocabulary.

iii. Regular Expressions

iv. NLTK package

v. Different models under NLP like BOW, CBOW, Word Embedding Model,TF-idf, word2vec

7. Some basic libraries you should know Pandas, Numpy, Matplotlib, Scikit, sklearn, tensorflow.

8. Basics of RDBMS, SQL commands like joins, modify any row or columns etc .

9. Concepts of statistics and some terminologies like Probability Distribution, Sampling, Hypothesis, p-value, bayesian statistics.

10. Be ready with your projects that you had mentioned in your resume. Interviewer can ask to show your project outputs, graphs. Make sure to have all your codes and outputs on your github account.

11. There are less chance of practical, but in some cases interviewer can ask to share your screen and give some questions with some time limits on Google colab or any other platform, so be prepared about it.

“Torture the data, and it will confess to anything.” — Ronald Coase

Connect with me on LinkedIn and Github.

--

--