- Explain how BN/LN works in NN - How does CNN works - What is transformer architecture
Sr Data Scientist Interview Questions
3,389 sr data scientist interview questions shared by candidates
Comparision of linear and non-linear models, evaluation metrics, high-level concepts on data mining algos.
Normal interview: business problem, classifier etc ... then he started with the obscure stuff such as pythons GIL and the f1beta score.
1-SQL aggregate functions with sorting 2- finding prime numbers 3- finding even numbers 4- differences in an array 5- dataframe
Problem Description: You are standing on a unique diamond-shaped platform composed of 5 tiles. These tiles are positioned at coordinates: (-1,0), (0,-1), (0,0), (0,1), and (1,0). From a starting position (xs,ys), you make random moves in one of four directions: left (decrease x by 1), right (increase x by 1), up (increase y by 1), or down (decrease y by 1). Each direction has the same probability and the direction of each move is entirely independent of previous moves. Determine the probability that you can reach a given destination (xe,ye) without stepping off the diamond platform. Constraints: (xs, ys) must be one of: (-1,0), (0,-1), (0,0), (0,1), (1,0) (xe, ye) must also be one of the above coordinates. Starting and ending coordinates are distinct: xs != xe or ys != ye Input: A single line containing four integers, denoting xs, ys, xe, ye respectively. Output: A single line showing the probability that you'll reach the destination before stepping off the platform. Sample input: -1 0 0 0 Sample Output: 0.25 Explanation: From the starting position, you have a 25% chance of moving right (and thus reaching the destination). Any other move would result in falling off the platform. [execution time limit] 4 seconds (py3) [memory limit] 1 GB [input] integer xs The x-coordinate of the starting position. [input] integer ys The y-coordinate of the starting position. [input] integer xe The x-coordinate of the end position. [input] integer ye The y-coordinate of the end position. [output] float Probability to reach end position before falling of platform.
1. Explain Matrix Factorization for the recommendation system. 2. Case study based question. 2. Explain generative adversarial networks. 3. Loss function of GAN 4. Coding question, solve n stairs problem. every time we can take step 1 ot step 2. print max possibles ways. use either dynamic or recursive approach.
Tipical HR interview, what do you know about Glovo etc etc
SQL, machine learning related questions
Focus on my projects, AWS knowledge, data science and ML basics
One of the questions at the panel interview focused on how to handle high-cardinality categorical variables in a model, especially when a naive approach could introduce noise or performance issues. The discussion was very practical and centered on trade-offs rather than theoretical tricks.
Viewing 2741 - 2750 interview questions
See Interview Questions for Similar Jobs
Jr Data ScientistData Scientist TraineeData Science EngineerSenior Machine Learning ScientistApplied Research ScientistSenior Data ScientistMachine Learning Research EngineerData Scientist NlpMachine Learning Research ScientistSenior Machine Learning EngineerData SciencesMachine Learning ScientistSoftware Engineer Machine Learning ScientistNlp Data ScientistResearch Scientist Computer VisionApplied ScientistSr. Data ScientistSenior Quantitative Researcher