I applied through a recruiter.

First was a standard recruiter screen where they mostly asked about my background and ML experience. After that, I had a 45-minute technical screen on CoderPad. It felt like a speed test, with two LeetCode medium problems and a pretty tight timer. There was nothing crazy conceptually, but you need to write clean code fast and keep explaining what you're doing. The onsite was a five-round loop:

• Two coding rounds
• One system design round
• One ML design round
• One behavioral round

The ML design round was probably the toughest. They asked me to design something like a large-scale recommendation system for a news feed and went pretty deep into data, features, models, and evaluation. The system design round was more traditional infrastructure stuff, such as scaling and distributed systems. The behavioral round was again standard, mostly focusing on past experience and handling ambiguity. One thing I realized, though, is that they want specific signals and features in depth, not vague answers. I found a good Meta ML coach on Prepfully for the mock, and he helped with the ML design part of the preparation. I'd say get a mock, be it on any platform, but get it. That helps a lot more than we think!

Design a recommendation system for a news feed.