Diagnocat
Technical interview at Diagnocat
A real interview with the full timeline, linked technical questions and practice materials from Diagnocat.
4 questions57 minAudio recording available
Аудио и материалы
Аудио собеседования
0:00 / 56:31
Этап 4 из 4DiagnocatML Engineer, медицинское CV2025-10-03 - 2025-10-22
Собеседования в Diagnocat: ML Engineer, медицинское CVТехническое собеседование в Diagnocat
Выводы и как готовиться
- The main signal is production ML hygiene: contracts between Dataset, collate, model, loss and optimizer.
- The interviewers push beyond “make it run” into responsibility boundaries: Dataset should not own GPU transfer or training state.
- Python runtime traps matter in ML code too: parent-class initialization and mutable default arguments can silently break training code.