Nvidia Interview Experience for QA SDET Intern (On-Campus)
Interview process held in Nov 2021 for internship duration of 5 months starting from Jan to Jun. Hackerrank test consists of 2 coding problems, OS, computer fundamental, and logical reasoning.
Shortlisted about 16-20 students. The interview duration was 1 Hr. there were two interviewers on MS Teams. One was asking CPP questions and the other was asking python
- Questions were based on Project
- What is the difference between lemming vs stemming
- What are things from NLP you used?
- Do you know any text classification algorithms?
- Define a string of length l in python
- Tell me about dynamic memory allocation
- What is the difference between LIFO vs FIFO
- Create a new list of words from the given list where the substring ‘ant’ is present in the word
- Asked me to code for max min element from the array
- There were a question about hardware
- What is the difference between SSD and HDD?
- Types of SSD?
- Tell me CPU parts.
- How computer boots, bios?
- What is blod, and how did it has occurred?
- Which games you played did yed what game settings you change (FPS, resolution)
- How to disable/enable storage devices from bios?
- How to block any service or app at startup
- They asked quean stion on ML also as there was opening for ML tool developer.
- Working and equation of SVM regressor ?
- Situation-based q on ML algorithm to choose.
- What is convolution ?
- Difference between logistic and linear regression
- Whado t is neural network
- Which graphics card you know which is latest GPU
- Hopefully I was able to answer the moan and st of the questions and I got the offer for internship.
- IMP topics to study
- Davisualizationion and data cleaning
- Different types of ML models and how they work
For Eg: regression and types of regression and how they work (algorithm)
- If you have studied DNNs then
Back and forward propagation, model training, neurons, and DNN layers, gradient descent algorithm, cost optimization
These are just the basic things expected to have
- Then you have your model analysis part
Errors (rmse, r-squared, least squares, etc)
Then there is model metrics (accuracy, precision, etc)
This much if you can cover it would cover most the things that can be asked