Turing Data Analyst interview experience

In the coding part, I had to solve two questions in Python.

After that, I moved into assessing an LLM model response they provided. I had to judge what was good or wrong with the output and evaluate it against the question context, which felt more like reviewing reasoning than just writing code.

Overall it was an average-difficulty experience for me, and while I completed the steps, I didn’t receive an offer. The model-output analysis was the part that made the process feel distinct from a purely coding interview.