I had an online HackerRank/LeetCode-style prompt under a tight window, immediately followed by a full-stack repository debugging task. An API call did not behave as expected, and I had to diagnose and patch it until every test passed.
The difficulty came from the number of moving pieces rather than the algorithm alone. In versions of this assessment style, there was also a work simulation where I used AI to solve a website bug, plus multiple behavioral OA questions. It was difficult and immersive, requiring sustained focus while making the pass condition clear. I did not receive an offer, and felt it measured messy, realistic failure modes rather than textbook problems.
Location: Jerusalem, Jerusalem. Overall feedback: neutral. Offer status: no offer.
Discussion
Loading comments…