What is Optuna and what are its pruners?
Machine Learning / AI
How to process tables and images in documents for RAG?
What validation methods for time series models exclude looking into the future during training?
What is speculative decoding and how does it speed up generation?
What pitfalls can occur when normalizing data to solve the extrapolation problem in a random forest?
How does the BERT model differ from GPT in architecture and training task?
What are adversarial examples and FGSM attack?
Design a multi-tenant LLM service with rate limits and subscription quotas.
Tell me about the most complex task or difficulty in one of your projects.
What is sample selection bias in an experiment platform?
What is self-ask prompting?
How to preprocess text: which stop words and punctuation to remove, use stemming or lemmatization?
Why is there no closed-form solution for logistic regression?
Does model drift occur during training with dropout and inference without it?
What problems can occur with one-dimensional positional encoding when working with images?
What is CUTLASS?
What protections against prompt injection do you know?
Describe the task of collapsing neighboring numbers into ranges?
How did you prepare documents for RAG (chunking, embeddings, Milvus)?
How to interpret the results of a linear regression model?