The day before the interview the interviewer was changed. The new interviewer had only been at the company for 4 months (I don't understand how that person is allowed to interview anyone). I had to share my screen. We were required to use HackerRank's AI, with agentic enabled. At the start he wasted 5-7 minutes picking the environment (Python) for me, and I don't know why he wouldn't let me pick it myself.
The problem was the expense rule problem from this forum, copied directly. The last sentence of the problem says it needs to support adding rules through an API later.
My approach:
- Clarify requirements
- Plan mode, sketch the main idea and structure
- Review the plan and execute
- Review the code
While I was reviewing the plan, he changed his mind several times. At first he said OK to an operand + operator approach (I had seen a post on the forum saying that hardcoding is an instant fail, so I never intended to hardcode), and then he suddenly told me to hardcode the evaluation function, so I had to change it twice.
At the end we were supposed to discuss scale, but we were already out of time. The AI-generated code also had some import issues, so it never ran. We just moved on. A very chaotic interview experience.
Discussion
Loading comments…