Comcast India Engineering Center Development Engineer 3 Interview Questions and Answers

Find median of 2 sorted arrays in O(log N) time complexity and O(1) space complexity

• Use binary search to find the partition point in both arrays

• Calculate the median based on the partition point and array sizes

• Adjust the partition points based on the median value

• Repeat until the partition points are at the median

• Handle edge cases such as empty arrays and uneven array sizes

Anagram of strings in O(1) space complexity

• Use a fixed size array of integers to store the frequency of characters in the first string

• Iterate through the second string and decrement the frequency of each character in the array

• If all the frequencies are zero, then the strings are anagrams

• Return true or false accordingly

Level order traversal of a tree using Queue

• Create a queue and add the root node to it

• While the queue is not empty, remove the front node and print its value

• Add the left and right child nodes of the removed node to the queue

• Repeat until the queue is empty

Reverse level order traversal of a tree using Queue

• Create a queue and push the root node into it

• While the queue is not empty, pop the front node and push its children into the queue

• Add the popped node to a stack

• Once the queue is empty, pop elements from the stack and print them

BFS and DFS are graph traversal algorithms. BFS explores nodes level by level while DFS explores nodes depth by depth.

• BFS uses a queue while DFS uses a stack or recursion.

• BFS is optimal for finding shortest path while DFS is optimal for finding a path between two nodes.

• BFS requires more memory as it stores all the nodes at each level while DFS requires less memory.

• BFS can be used to find connected components while DFS

Find product of each element of an array except that element in O(N) time complexity without using / operation

• Use prefix and suffix products

• Multiply prefix and suffix products for each element to get the final product

• Handle edge cases where array has 0 or 1 element separately

Recursively delete a linked list

• Create a recursive function that takes the head of the linked list as input

• Base case: if the head is null, return

• Recursively call the function with the next node as input

• Delete the current node

Recursively delete a linked list from the end.

• Start from the head and recursively traverse to the end of the list.

• Delete the last node and set the second last node's next pointer to null.

• Repeat until the entire list is deleted.

• Use a recursive function to implement the deletion process.

Recursively delete a tree by deleting all its child nodes and then the parent node.

• Start from the leaf nodes and delete them first.

• Then move up to the parent nodes and delete them.

• Repeat until the root node is deleted.

• Use post-order traversal to ensure child nodes are deleted before parent nodes.

Recursively delete elements from the end of an array.

• Create a recursive function that removes the last element of the array.

• Call the function recursively until the desired number of elements are removed.

• Handle edge cases such as empty arrays and removing more elements than the array contains.

Floyd Warshall finds shortest path between all pairs of vertices while Djikstra finds shortest path from a single source.

• Floyd Warshall is used for dense graphs while Djikstra is used for sparse graphs.

• Floyd Warshall has a time complexity of O(n^3) while Djikstra has a time complexity of O((n+m)logn).

• Floyd Warshall can handle negative edge weights while Djikstra cannot.

• Floyd Warshall can detect negative cycles while Dj

There is no known algorithm to find shortest path in 2-D space in O(log N) time.

• The best known algorithm for finding shortest path in 2-D space is Dijkstra's algorithm which has a time complexity of O(N^2).

• Other algorithms like A* and Bellman-Ford have better time complexity but still not O(log N).

• If the points are on a grid, Lee algorithm can be used which has a time complexity of O(N).

Design a system for putting newspapers using classes and functions

• Create a Newspaper class with attributes like title, date, and content

• Create a Publisher class with methods to publish and distribute newspapers

• Create a Subscriber class with methods to subscribe and receive newspapers

• Use inheritance to create different types of newspapers like daily, weekly, etc.

• Implement a database to store newspaper information and ha

I believe practical experience is more valuable than higher studies.

• I have gained valuable experience through internships and projects.

• I prefer hands-on learning and problem-solving over theoretical knowledge.

• I am constantly learning and improving my skills through online courses and workshops.

Detect if two circular linked lists overlap

• Traverse both lists and check if they have the same tail node

• If they have different tail nodes, they do not overlap

• If they have the same tail node, check if they intersect at any point

• Use Floyd's cycle-finding algorithm to detect intersection point