Software Engineer · Online Assessment · Coding

IBM — Software Engineer ✅ Passed

Level: Junior-Level

Round: Online Assessment · Type: Coding · Difficulty: 5/10 · Duration: 60 min · Interviewer: Neutral

Topics: Dynamic Programming, Greedy Algorithms, Prefix Sum, Sliding Window

Location: San Francisco Bay Area

Interview date: 2025-06-06

Summary

I completed an online assessment with two questions.

Details

I had to solve two problems:

getMinTime:

The problem describes 'n' features that need to be unlocked and deployed in sequence. Each feature has an initial deployment cost and a repeated deployment cost. I needed to calculate the minimum total time to complete a specified number of total deployments.

My approach involved iterating through all possible strategies. A strategy involves determining how many of the initial features to unlock and then using the remaining deployment counts on the cheapest repeated deployment. I used an O(n) loop, maintaining two variables in real-time: the 'cumulative unlock cost' up to the current feature (a prefix sum) and the 'minimum repeated deployment cost' up to the current feature (a sliding minimum). In each step, I calculated the total time for a strategy of 'unlocking up to this point', then compared it with the global minimum and updated accordingly.

While there isn't a direct LeetCode equivalent, this problem tests a simplified application of dynamic programming (DP) or a strategy-based greedy approach, utilizing prefix sum and sliding minimum techniques.

I used BigInt to prevent overflow, as the value ranges were large. Some test cases seemed counterintuitive or different from initial calculations, affecting the runtime.

maximizeDecryptionValue:

This problem involves array processing. The prompt describes files, some encrypted and some decrypted, each with a score. I could perform one operation: select a contiguous range of length no more than 'K' and decrypt all files within it. The goal was to maximize the total score of all decrypted files.

My solution split this into two parts: 1. The sum of scores of already decrypted files (base score). 2. The maximum 'additional' score obtainable via one operation. This converted the problem into finding the maximum sum of a contiguous subarray of length no more than 'k' in an array.

This is a sliding window problem, similar to LeetCode 209. Minimum Size Subarray Sum. The approach involves maintaining a window of variable size to solve it in O(n).

LeetCode similar: LeetCode 209