U R Rao Satellite Centre Research Scientist Interview Questions and Answers

Coding round 2 questions

Q1. Asked Algorithms used in the project (No follow-up on mentioned algorithms, cut-off mid-explanation of business problem, scale, and solution wanting to know just the names of the algorithm) - Answered by n...read more
Add your answer
Q2. Count all pairs of numbers from a list where the ending digit of the ith number equals the starting digit of the jth number. Example [122, 21, 21, 23] should have 5 pairs (122, 21), (122, 21), (122, 23), (...read more
Ans. 
Count pairs of numbers where ending digit of ith number equals starting digit of jth number.
• Iterate through each pair of numbers in the list

• Check if the ending digit of the ith number equals the starting digit of the jth number

• Increment the count if the condition is met
Answered by AI
Add your answer
Q3. Interpretation of graphs, the first graph had perpendicular lines from the error to the fitted line and the second graph had lines from the error to the fitted line, parallel to the y-axis. - Interpreted t...read more
Ans. 
Interpretation of graphs in linear regression analysis
• Perpendicular lines from error to fitted line in first graph indicate OLS using projection matrices

• Lines parallel to y-axis from error to fitted line in second graph suggest evaluation of linear regression to y-pred - y-actual method

• PCA could also be a possible interpretation for the second graph
Answered by AI
Add your answer
Q4. What does np.einsum() do
Ans. 
np.einsum() performs Einstein summation on arrays.
• Performs summation over specified indices

• Can also perform other operations like multiplication, contraction, etc.

• Syntax: np.einsum(subscripts, *operands)
Answered by AI
Add your answer
Q5. How to generate random numbers using numpy, what is the difference between numpy.random.rand and numpy.random.randn
Ans. 
numpy.random.rand generates random numbers from a uniform distribution, while numpy.random.randn generates random numbers from a standard normal distribution.
• numpy.random.rand generates random numbers from a uniform distribution between 0 and 1.

• numpy.random.randn generates random numbers from a standard normal distribution with mean 0 and standard deviation 1.

• Example: np.random.rand(3, 2) will generate a 3x2 array of r...read more
Answered by AI
Add your answer
Q6. Difference between logit and probabilities in deep learning
Ans. 
Logit is the log-odds of the probability, while probabilities are the actual probabilities of an event occurring.
• Logit is the natural logarithm of the odds ratio, used in logistic regression.

• Probabilities are the actual likelihood of an event occurring, ranging from 0 to 1.

• In deep learning, logit values are transformed into probabilities using a softmax function.

• Logit values can be negative or positive, while probabili
Answered by AI
Add your answer

Q1. Basics of linear regression
Add your answer

Q1. Machine learning in depth
Add your answer

Q1. Difference between bagging and boosting
Ans. 
Bagging and boosting are ensemble learning techniques used to improve the performance of machine learning models by combining multiple weak learners.
• Bagging (Bootstrap Aggregating) involves training multiple models independently on different subsets of the training data and then combining their predictions through averaging or voting.

• Boosting involves training multiple models sequentially, where each subsequent model c...read more
Answered by AI
Add your answer
Q2. Parameters of Decision Tree
Ans. 
Parameters of a Decision Tree include max depth, min samples split, criterion, and splitter.
• Max depth: maximum depth of the tree

• Min samples split: minimum number of samples required to split an internal node

• Criterion: function to measure the quality of a split (e.g. 'gini' or 'entropy')

• Splitter: strategy used to choose the split at each node (e.g. 'best' or 'random')
Answered by AI
Add your answer
Q3. Explain any one of your project in detail
Ans. 
Developed a predictive model to forecast customer churn in a telecom company
• Collected and cleaned customer data including usage patterns and demographics

• Used machine learning algorithms such as logistic regression and random forest to build the model

• Evaluated model performance using metrics like accuracy, precision, and recall

• Provided actionable insights to the company to reduce customer churn rate
Answered by AI
Add your answer

NER training using deep learning

Q1. Describe the approach taken for assignment
Ans. 
I approach assignments by breaking them down into smaller tasks, setting deadlines, and regularly checking progress.
• Break down the assignment into smaller tasks to make it more manageable

• Set deadlines for each task to stay on track

• Regularly check progress to ensure everything is on schedule

• Seek feedback from colleagues or supervisors to improve the quality of work
Answered by AI
Add your answer
Q2. Scenario based questions
Add your answer

Q1. Overfitting and Underfitting
Add your answer
Q2. Find Nth-largest element
Ans. 
Find Nth-largest element in an array
• Sort the array in descending order

• Return the element at index N-1
Answered by AI
Add your answer
Q3. NLP Data preprocessing
Add your answer

Q1. Salary Discussion
Add your answer
Q2. Fitment discussion
Add your answer

Q1. Which GenAI projects I have worked on
Add your answer
Q2. What is the context window in LLMs
Ans. 
Context window in LLMs refers to the number of surrounding words considered when predicting the next word in a sequence.
• Context window helps LLMs capture dependencies between words in a sentence.

• A larger context window allows the model to consider more context but may lead to increased computational complexity.

• For example, in a context window of 2, the model considers 2 words before and 2 words after the target word fo
Answered by AI
Add your answer
Q3. What is top_k parameter
Ans. 
top_k parameter is used to specify the number of top elements to be returned in a result set.
• top_k parameter is commonly used in machine learning algorithms to limit the number of predictions or recommendations.

• For example, in recommendation systems, setting top_k=5 will return the top 5 recommended items for a user.

• In natural language processing tasks, top_k can be used to limit the number of possible next words in a
Answered by AI
Add your answer
Q4. What are regex patterns in python
Ans. 
Regex patterns in Python are sequences of characters that define a search pattern.
• Regex patterns are used for pattern matching and searching in strings.

• They are created using the 're' module in Python.

• Examples of regex patterns include searching for email addresses, phone numbers, or specific words in a text.
Answered by AI
Add your answer
Q5. What are iterators and tuples
Ans. 
Iterators are objects that allow iteration over a sequence of elements. Tuples are immutable sequences of elements.
• Iterators are used to loop through elements in a collection, like lists or dictionaries

• Tuples are similar to lists but are immutable, meaning their elements cannot be changed

• Example of iterator: for item in list: print(item)

• Example of tuple: my_tuple = (1, 2, 3)
Answered by AI
Add your answer
Q6. Do I have REST API experience
Ans. 
Yes, I have experience working with REST APIs in various projects.
• Developed RESTful APIs using Python Flask framework

• Consumed REST APIs in data analysis projects using requests library

• Used Postman for testing and debugging REST APIs
Answered by AI
Add your answer