What is differential privacy and DP-SGD?
Machine Learning / AI
What is EAGLE speculative sampling?
How to calculate cosine similarity between questions?
What is Dreambooth and how to avoid overfitting?
What streaming inference techniques for STT (chunked attention, look-ahead)?
What is Dice Loss and why is it often better than BCE for medical segmentation?
What are anchor-free detectors? Tell us about FCOS.
What does your development process look like — do you bring models to Docker, wrap them, and pass them on to someone else?
What is Inverse Propensity Scoring in LTR?
What techniques do you use to prevent overfitting in boosting (subsample, colsample, min_child)?
Tell about the forecasting task from your resume — what is the model built on and what was the prediction for?
What is pivoted normalization?
What is compute-optimal training according to Chinchilla?
What is Apache Beam / Dataflow?
How to use collaborative methods for recommendations based on embeddings?
What is agent sandboxing and why is it needed?
What is ontology engineering and its role in KG?
How do you evaluate an unsupervised model for a question-answering task?
What is Snapdragon AI Engine?
What data types in Python are mutable among list, set, dict?