razorpay logo
Razorpay· Senior Machine Learning Engineer

Razorpay | OA (1st Round) | Senior MLE | Bengaluru

Location
Bengaluru
Experience
4 years
Posted
26 Jun 2025
433 reads

Summary

I recently completed an Online Assessment at Razorpay in June 2025 for a Senior Machine Learning Engineer role. The test consisted of three algorithmic problems, including LFU cache implementation, a string partitioning challenge, and a knapsack variant.

How it went

Recently I gave Online Assesment at Razorpay in June 2025 and these were the questions which were asked. I have 4 years of experience as a data scientist, the role was Senior Machine Learning Engineer.

Platform: HackerEarth
Mode: Online Coding Test
Round: 1 (Online Assessment)
No. of Questions: 3
Duration: 90 Minutes
Difficulty: Medium to Hard

Questions asked (3)

  1. 01

    LFU Cache Implementation

    Data Structures & Algorithms

    Implement a cache using the Least Frequently Used (LFU) eviction policy. When the cache reaches its capacity, remove the least frequently used key. If there's a tie, remove the smallest key among those with the least frequency.

    Function Signature:

    def solve(N: int, Q: int, operations: List[List[int]]) -> List[int]
    

    Parameters:

    • N: Capacity of the cache
    • Q: Number of operations
    • operations: List of operations where:
      • Type 1: 1 key → Get value from cache
      • Type 2: 2 key value → Insert/Update key-value pair

    Constraints:

    • 1 ≤ N ≤ 5 * 10^4
    • 1 ≤ Q ≤ 2 * 10^5
    • 1 ≤ key ≤ 2 * 10^5
    • 1 ≤ value ≤ 10^9

    Sample Input:

    2
    5
    1 2 -1
    2 1 3
    2 2 4
    2 4 5
    1 2 -1
    

    Sample Output:

    -1 4
    
  2. 02

    Minimum Points Partition of String

    Data Structures & Algorithms

    You're given a string S. Split it into parts such that each character appears in at most one part. The score of a part is the square of its length. Minimize the total score across all parts.

    Function Signature:

    def solve(S: str) -> int
    

    Scoring Rule:

    • For part of length x, score = x^2

    Constraints:

    • 1 ≤ len(S) ≤ 500
    • S[i] contains lowercase letters

    Sample Input:

    eccbbbbdec
    

    Sample Output:

    100
    
  3. 03

    Min Cost with Constraints (Knapsack Variant)

    Data Structures & Algorithms

    You're given N types of chip packets, each with:

    • happiness A[i]
    • weight B[i]
    • cost C[i]

    Alice wants at least X happiness and at most Y total weight while minimizing the cost.

    Function Signature:

    def solve(N: int, A: List[int], B: List[int], C: List[int], X: int, Y: int) -> int
    

    Parameters:

    • A: Happiness values (length N)
    • B: Weights (length N)
    • C: Costs (length N)
    • X: Minimum required happiness
    • Y: Maximum allowed total weight

    Constraints:

    • 1 ≤ N ≤ 100
    • 1 ≤ A[i], B[i], C[i], X, Y ≤ 100
    • 1 ≤ C[i] ≤ 10^4

    Sample Input:

    5
    1 2 3 4 5
    1 2 3 4 5
    5 4 3 2 1
    10
    10
    

    Sample Output:

    5
    

How they prepared

The test had a mix of system design logic (LFU cache), greedy optimization, and classic DP (knapsack). All questions were algorithmically rich and tested core CS fundamentals. Strong understanding of hash maps, priority queues, and dynamic programming is essential for clearing this round.

Know someone preparing? Send it to them:WhatsAppLinkedInX

Interviewed at Razorpay recently?

Write up your rounds while you still remember the questions. It helps the next person more than you would think.

Share your experience
Discussion 0

Questions for the author, or how your own interview compared.

Join the discussion

Sign in with Google to comment or ask the author a question.

No comments yet

Be the first to share your thoughts and start the discussion!