Applied through Amazon jobs portal.
Number of rounds – 5
Overall time taken –1.5 month.
YOE - 1.5 year
Given a List of Orders , where each Order is a string . For example : [aax 12 23 ] [ ff kindle ebook] are two orders. Each order has an ID which is first part of the order ( ID of order 1 = aax , ID2 = ff ). The remaining part of the order is referred to as MetaData. There are two types of orders, Prime orders ( which contain non numeric Meta Data) and Non-Prime Orders (which contain Only-Numeric Meta Data). Sort the List of Orders as given below :
a. Prime Orders before NonPrime Orders
b. Prime Order with lexicographically smaller MetaData comes first.
c. In Case of tie in (b) , the one with lexicographically smaller ID comes first.
d. The relative order of NonPrime Orders remains the same.
Passed all test case.
A combo is defined as a subset of the given n terms. The total popularity is the sum of the individual items of the combo. design an algorithmn that can find the k combos with the highest popularity.
Two combos are considered different if they have different subset of items. Return the array of k integers where the ith term denotes the popularity of ith best combo. Combos should be returned arranged best to worst.
Note: You can have an empty subset as a combo as well. The popularity for such a subset is 0.
Example
n = 3
popularity= [3, 5, -2]
k = 3
All possible popularities of combos are 0, 3, 5, -2, 8, 3, 1, 6.
The best three combos have popularities 8, 6, and 5. The answer is [8, 6, 5].
Returns
int[k]: the popularities of the best k combos, in decreasing order.
Constraints
1 <= n <= 10^5
-10^9 <= popularity[i] <= 10^9
1 <= k <= min(2000, 2^n)
Passed 11/15 test case.
Orginally Interviews were scheduled for 3rd week of December but got postponed due to year end. Therefore, it happened during the second week of January
Short Introduction
Remove nodes on root to leaf paths whose length < K
Hard level problem
Largest Area in Histogram.
Hard level problem
Explained: Brute{ O(n)² } -> Better{ 3 pass O(n) } -> Optimal{ 1 pass O(n) }
LP.
Short Introduction
Find the Next Greater Element in an array.
Medium level problem
First solve with Brute force approach. Then optimize it with Stack taking O(n) space.
Convert a Ternary Expression string to a Binary Tree.
Here you can have a string instead of char between the ternary operators.
Hard level problem
Input : ab ? bcd : efgh
Output
ab
/ \
bcd efgh
LP.
Introduction
Convert 1 into N in minimum steps by multiplying with 2 or by adding 1.
Input: 19; Output: 6
Medium level problem
Explained: Recursion -> DP -> Better{ O(n) } -> Optimal{ O(log(n)} }
After that question was updated with if you are only allowed to multiply by 2, K times.
Explained: Optimal{ O(logK) }
LP and Behaviour: Took a different approach where I asked permission for presenting myself, then through my presentation only all his/her questions were answered automatically.
Introduction
It contained Leadership - Behaviour - Situtation based questions. (Can't remember exactly, as it was more of a conversation rather than a typical QA round)
Verdict – Selected.