Nykaa MLS2 / Data Scientist 2 Interview Experience

nykaa logo
nykaa
· Data Scientist 2· Bangalore, India· 5y exp
May 5, 2026 · 0 reads

Summary

Cleared initial rounds including coding, statistics, and machine learning questions but was ultimately rejected during the hiring manager round due to perceived average performance and compensation negotiation.

Full Experience

Background: Leading Global B2C Company, Gurgaon
Experience: 5 Years
Location: Bangalore, India
Current: 31 LPA
Inital Ask: 40 LPA

How I Applied

I was contacted by a third party recruiter through Instahyre. After a brief discussion, I was shortlisted for the interview process.

Round 1 – Data Structures, Coding & Statistics (1 Hr)
Coding Questions
1. Implement Mean, Median, and Mode in core Python without using library functions.
2. Implement a Decision Tree from scratch in Python.
I was not able to complete the Gini Impurity logic but explained concepts like threshold selection, depth, and other parameters.

Probability Questions
. Given a biased coin, how would you generate unbiased outcomes?
. If a 1-meter stick is broken into 3 random parts, what is the probability that the parts can form a triangle?

✅ Verdict: Selected

Round 2 – Hands-on ML Coding (1 Hr)
Asked to perform operations like GroupBy and Window functions on a dataset using PySpark. (I requested to solve it using Pandas/SQL due to syntax not in memory, and the interviewer agreed, However he was not too happy about it)
Implement Linear Regression from scratch using:
Core Python, Pytorch
Discussion Around
. Data leakage
. Feature engineering
. Lag features in time-series forecasting

✅ Verdict: Selected
(Received feedback later that my performance was considered average.)

Round 3 – ML/DL Breadth & Domain Expertise (30–40 Mins)
Time-series forecasting and different forecasting models
Cold-start problem in forecasting
Zero-sales/dead article forecasting problem
Resume and project deep dive

The interviewer validated my experience and projects thoroughly, and I was able to answer all project-related questions confidently.

✅ Verdict: Selected

Round 4 – ML Case Studies & Fitment (Hiring Manager) (1 Hr 10 Mins)
He went over the previous discussions, shared the interview feedback with me, and then asked why I thought the interviewer had given that feedback, especially regarding the areas they considered comparatively weaker.

Statistics Questions
What is a Random Variable, and what is random about it?
I misunderstood the intent and explained random initialization in neural networks instead of the probability perspective.
Explain any distribution other than the Normal Distribution and its applications.
→ Discussed Student’s t-distribution.
What is the Central Limit Theorem?
I could explain its applications but could not clearly recall the exact theorem statement.
What is Bias, and why is it called so in the Bias-Variance tradeoff?
Other Topics
Resume and project discussions
Basic MLOps concepts like Redis and Kafka and how they were used in projects, Real time latency

The HM was quite transparent about the company culture, mentioning:
. High ownership expectations and initatives
. Upon asking, he mentioned limited benefits: no lunch/snacks, no cab service, little work-from-home, and limited flexibility in timings

At the same time, he did said that company values employees and do not fire often.

❓ Verdict: Got call from company HR, asking how was interview and expectation, I asked for 45 LPA (plus relocation), but the HR team seemed hesitant and inital HR upon calling for confirmation informed No hire and Rejected.

Update: Work culture is a major factor for me, and only higher compensation could offset that concern. So I decided to not go any lower. Since I am not a frequent switcher, I would prefer to wait for an MNC offer.

Interview Questions (9)

1.

Implement Mean, Median, and Mode in Core Python

Data Structures & Algorithms·Medium

Write code to compute mean, median, and mode of a list of numbers using only core Python, without importing any external libraries such as numpy or statistics.

2.

Implement Decision Tree from Scratch

Data Structures & Algorithms·Hard

Build a decision tree classifier from scratch using pure Python. Include node splitting based on Gini impurity or information gain.

3.

Unbiased Outcome Using Biased Coin

Data Structures & Algorithms·Medium

Given a biased coin which may not land heads/tails with equal probability, describe how you can generate fair/unbiased outcomes using this coin.

4.

Triangle Formation Probability from Stick Breaking

Data Structures & Algorithms·Hard

A stick of length 1 meter is randomly broken into three pieces. What is the probability that these three pieces can form a triangle?

5.

Linear Regression from Scratch

Data Structures & Algorithms·Medium

Implement linear regression model from scratch using only core Python or PyTorch tensors, without using sklearn or similar frameworks.

6.

Random Variable Definition

Behavioral

Define what a Random Variable is and clarify what makes it 'random' in mathematical terms.

7.

Student’s t-Distribution Explanation

Behavioral

Explain Student’s t-distribution and provide examples where it is more appropriate than normal distribution.

8.

Central Limit Theorem Statement

Behavioral

State and explain the Central Limit Theorem and its practical implications in statistical inference.

9.

Bias in Bias-Variance Tradeoff

Behavioral

Explain the concept of bias in machine learning models and how it contributes to the overall error along with variance.

Preparation Tips

Focused on implementing foundational algorithms in Python, practiced statistics basics, reviewed ML theory and applied ML case studies. Also brushed up on system design aspects related to real-time systems and MLOps tools like Redis and Kafka.

📣 Found this helpful? Please share it with friends who are preparing for interviews!

Discussion (0)

Share your thoughts and ask questions

Join the Discussion

Sign in with Google to share your thoughts and ask questions

No comments yet

Be the first to share your thoughts and start the discussion!