Aurélien's ability to simplify complex Machine Learning concepts, reminiscent of the Feynman study method (productive.fish/blog/feynman-learning-technique), was the highlight of this interview for me. His emphasis on selecting the right metrics in ML projects and the challenges of labeling are invaluable insights for anyone in the field. The discussion underlines the importance of continuous learning in AI and ML, a crucial trait for any practitioner. This interview is not only enlightening but also inspiring for continuous growth in this dynamic field. Great read!