I went through full interview loop at Google for L6 MLE position. This is how it went. I won't be able to share the exact questions.

Round1: ML system design, went well. Question was about designing a content moderation system. Feedback: Positive

Round2: ML system design, went very well. Question was about building a system which deals with traffic data. Feedback: Positive

Round3: Coding, I was asked a question similar to logger timer, but the interviewer kept adding more constraints. I was able to solve all of them. Seemed like the interviewer was happy. Feedback: Positive

Round4: Coding, I was aksed a question based on a Grid. Initially I proposed a O(n^2) solution and pro-actively optimised to O(n), implemented the solution. Interviewer then asked a slightly harder follow up for which I proposed a brute-force approach, which the interviewer asked me to implement. I ran out of time before I could finish the implementation, I had implemented maybe 2/3rds of it, and I hope I communicated my idea well. Feedback: Negative. [This interview was supposed to be an ML coding round, but the interviewer did not ask any ML related questions.]

Round5: Googleyness & Leadership, this interview went very well in my reckoning. The interviewer seemed happy, we chit chatted beyond the interview time about Google's culuture etc. Feedback: Positive.

Update: Received official feedback today. Recruiter said we're in a decent position, and will move to team-matching stage.

Resources for preparation:

Machine Learning System Design:

1. This is a good concise repo of various ML design problems: https://github.com/alirezadir/Machine-Learning-Interviews/blob/main/src/MLSD/ml-system-design.md
2. A good advice on ML design by Patrick Halina: http://patrickhalina.com/posts/ml-systems-design-interview-guide/#leveling
3. A good article by Bharati Priya: https://medium.com/@reachpriyaa/how-to-crack-machine-learning-interviews-at-faang-78a2882a05c5
4. A huge collection of blogs on ML systems: https://www.evidentlyai.com/ml-system-design
5. Rules of ML by Google : https://developers.google.com/machine-learning/guides/rules-of-ml. This is a very good resource regardless of whether you're applying for Google.

ML Coding:

1. Neetcode has a few ML coding exercises: https://neetcode.io/practice
2. A good collection of ML coding challenges: https://www.deep-ml.com/?page=1&difficulty=&category=&solved=
3. A great article by Abhijit on ML coding problems: https://mecha-mind.medium.com/ds-algo-problems-ml-coding-878b852ea421

Behavioural:
Mostly past experiences. Use chatGpt for simulating the interview experience. You can refer to this link, might be helpful. https://www.themuse.com/advice/behavioral-interview-questions-answers-examples