The interview went well overall, though the salary offered was slightly below expectations, making the final decision a bit difficult. The technical rounds focused heavily on classical machine learning concepts, testing foundational knowledge such as:
Bias-variance tradeoff
Regularization techniques
Feature engineering and selection
Evaluation metrics (precision, recall, F1, ROC-AUC)
Basic optimization and gradient descent intuitions
In addition to classic ML, the interview also included in-depth questions around LLMs and Transformer architectures. To perform well, a thorough understanding of the following was essential:
