Overall a good but demanding process. It consisted of 7 rounds:

1. Recruiter call
2. Technical call (Object Oriented Programming)
3. Manager call
4. Last round: Virtual onsite with 4 back-to-back interviews: a. Technical (OOP) b. Behavioral c. Applied ML d. ML Fundamentals

The interview process is clear, and they use ModernLoop to schedule interviews and prepare for them. It is very time-demanding. I did over 7 hours of interviews in total.

The ML Fundamentals round involves reading a research paper and answering questions. Be sure to please your interviewer and be highly optimistic about LLM research and AI research. I stated that ML research was not fundamental research, and I do not think the interviewer liked that.

I was denied for "not performing well" at the ML Fundamentals interview. I am obviously very disappointed after dedicating so much time. I am also reassured not to work with people who cannot accept different opinions and cannot review their own process. Outside the fact that it did not go well with one interviewer, if after doing well in all others, it does not seem strange, it is sad.

• Card game / Oriented Object Programming
• Simple RAG pipeline coding
• Review an ML research paper and answer questions
• What are my defaults and my qualities? How do they impact my work? What do I do about them? --> Classic behavioral questions