ML Debugging: Parse Conversation Data, Then Fix an LLM Project

Problem Overview

A 60-minute ML round in two parts, done in one session. First you write a simple parser for conversation data; then you are handed a complete LLM project, run it, and find and fix its bugs. This stand-in plants three. Part 1 — Parse the conversation data. Write parse_conversations(raw), which…

• The rest of the problem statement — full requirements, constraints, and edge cases
• Approach and trade-offs — what passing candidates did, and the mistakes that sink people