Navi Technologies interview experience Real time questions & tips from candidates to crack your interview
Data Scientist
Navi Technologies
3 rounds | 4 Coding
problems
Selected Interview preparation journey
Preparation
Duration: 4 months
Topics: ML algorithms - Decision trees, linear regression, logistic regression, PCA, ensemble methodsDeep Learning - ANN, CNN, RNN and their variations, theory behind themStatistics - Distributions, hypothesis testsPythonData structure and algorithms
Tip 1 : Practice > 200 DSA questions, Leetcode medium should be solvable 90% of times
Tip 2 : Be extremely clear with theoritical concepts related to machine learning and deep learning
Tip 3 : Have atleast 2 projects, about machine learning and deep learning
Application process
Where: Campus
Eligibility: No criteria for data science profile. SDE was only open for maths and CSE
Tip 1 : Have atleast 2 projects, ML and Deep learning
Tip 2 : Have atleast 1 internship
Interview rounds
01
Round
Medium
Online Coding Interview
Duration90 minutes
Interview date28 Oct 2022
Coding problem1
Round was scheduled in the evening in the institute, it was proctored by placement representatives.
1. Valid Sudoku
Moderate
40m average time
60% success
0/80
Asked in companies
You have been given a 9 X 9 2D matrix 'MATRIX' with some cells filled with digits(1 - 9), and some empty cells (denoted by 0).
You need to find whether there exists a way to fill all the empty cells with some digit(1 - 9) such that the final matrix is a valid Sudoku solution.
A Sudoku solution must satisfy all the following conditions-
1. Each of the digits 1 - 9 must occur exactly once in each row.
2. Each of the digits 1 - 9 must occur exactly once in each column.
3. Each of the digits 1 - 9 must occur exactly once in each of the 9, 3 x 3 sub-matrices of the matrix.
Note
1. There will always be a cell in the matrix which is empty.
2. The given initial matrix will always be consistent according to the rules mentioned in the problem statement.
Problem approach
Applied the basic rules of sudoku to check the validity
02
Round
Easy
Face to Face
Duration30 mins
Interview date1 Dec 2022
Coding problem1
was with the Director of engineering. The interview started with my introduction then he started asking questions about my projects. I had pinned the collab file in my browser and opened it, and started going over the code, explaining each line. Then he asked me if I had deployed it somewhere. To which I replied with a no, but I told him that it was one of my initial projects, that’s why I hadn't deployed it
After that, he asked basic ML questions like the hyperparameters in XGboost, as I had mentioned XGboost in my resume, and why I had used the algorithms, I had mentioned. In the end, he asked me if I had any questions. I advise always to ask something as it shows that you are genuinely interested in working for the company.
1. ML Questions
1) Explain KNN, decision trees
2) Explain XGBoost and what are it's hyperparameters
3) What sampling techniques were used to deal with imbalanced dataset
4) Overfitting and underfitting concepts
Problem approach
Tip 1: Prepare the theory portion of ML with depth
03
Round
Medium
Face to Face
Duration30 mins
Interview date1 Dec 2022
Coding problem2
This was also a technical round. The interviewer had a poker face during the whole interview, and it was tough to judge whether my answers were correct from his expressions. Without wasting any time, he directly started asking me questions. The round seemed to be emphasising more on the breadth of topics. The interviewer asked basic questions like loss function used in different classifiers, CLT, p-value etc. I answered almost all the questions satisfactorily.
At last, he gave me a problem on probability, which I guess I didn’t solve correctly during the interview. It was like - 60 people were standing in a row and divided into 3 groups of 20 people each randomly. Given two people, A and B, in the row. I was asked that probability that A and B come in the same group.
1. ML Questions
1) Loss function in Logistic regression
2) Statistical concepts like p-value, central limit theorem, hypothesis testing
3) Different optimizers used in deep learning
2. Puzzle
60 people were standing in a row and divided into 3 groups of 20 people each randomly. Given two people, A and B, in the row. I was asked that probability that A and B come in the same group.
Here's your problem of the day 
Solving this problem will increase your chance to get selected in this company
Skill covered: Programming
What is recursion?
Choose another skill to practice
Similar interview experiences
SDE - 2
3 rounds | 6 problems
Interviewed by Navi Technologies
Selected 
1074 views
0 comments
0 upvotes
SDE - 1
4 rounds | 8 problems
Interviewed by Amazon
Selected 8518 views
0 comments
0 upvotes
SDE - Intern
1 rounds | 3 problems
Interviewed by Amazon
Selected 3320 views
0 comments
0 upvotes
SDE - 2
4 rounds | 6 problems
Interviewed by Expedia Group
Selected 2580 views
0 comments
0 upvotes
Companies with similar interview experiences
Data Scientist
4 rounds | 8 problems
Interviewed by Walmart
Selected 2314 views
0 comments
0 upvotes
Data Scientist
2 rounds | 3 problems
Interviewed by Sprinklr
Selected 0 views
0 comments
0 upvotes
Data Scientist
2 rounds | 11 problems
Interviewed by Ernst & Young (EY)
Selected 2517 views
0 comments
0 upvotes