How to protect RAG from prompt injection and unwanted content?
Machine Learning / AI
How is coordination between LLM agents carried out?
What methods need to be implemented when inheriting from the Dataset class in PyTorch?
What tasks does ML solve in drones and cars?
What do you forget to do with data on CUDA that can lead to memory overflow during training?
If the optimal number of trees for a model is 500, but we accidentally trained both a random forest and boosting with 1000 trees, which model is more likely to overfit and why?
What examples of feature engineering have you performed in projects?
What is Pix2Struct?
How did you measure success in the [company] project (code autocompletion)? What metrics did you use?
What is the next method in iterators and how is it implemented?
What problems arise if batch normalization updates statistics during validation?
What to do if the model does not fit on the GPU during fine-tuning?
What types of data quality issues have you encountered?
What does a clustering task look like and what algorithms do you know?
What happens if you do select 1 instead of a column name?
What is CV-leak and what are its sources?
Tell us about your biggest task in the last year or two: context, impact on business, metrics.
How does the depth of a tree affect the model's quality and its generalization ability?
Was it somehow related to development or did you do something about it?
Tell us about your experience with the web frameworks Django and FastAPI?