Finished KP's HackerRank assessment last week and had the technical phone screen earlier this month. Posting data since the info out there is pretty thin.
Online assessment (HackerRank, 90 minutes): 2 coding problems Problem 1: medium difficulty, string manipulation, something you've seen in LeetCode medium practice Problem 2: medium-hard, graph traversal variant, nothing exotic but requires knowing BFS/DFS properly No time extensions in my experience, 90 min was enough if you're prepared No proctoring beyond the browser, standard HackerRank setup
Technical phone screen (45 min, Zoom): Not a pure coding screen. It was more of a technical conversation with one coding question. The question was roughly: given a list of patient appointment times, find conflicting slots. Not an exotic problem. They cared how I talked through it, not just whether I produced the solution. I was asked to explain my approach before writing a single line.
Difficulty overall: I'd call it LC medium. Nothing hard, nothing trivial. They're not trying to filter on algorithmic tricks, they want to see that you can write clean readable code and explain your thinking.
Language: I used Python, seemed totally fine. Saw another candidate in a Discord server who used Java.
One thing I didn't expect: during the phone screen they asked a few questions about testing. Not 'write unit tests' but more like 'how would you test this function, what edge cases matter?' That's the QA/SDET in me noticing, but it's real. KP seems to care about quality at the process level.
No SQL in my loop but I've heard it shows up for data-adjacent roles. For pure SWE it seemed algorithm + design focused.