What databases have you worked with?
Machine Learning / AI
Are you familiar with GigaChat? Do you have experience working with it?
What does the __init__ method do in a class and when is it called?
What methods of text chunking do you know?
What is the purpose of the 'g' object in LangGraph and why is it used in fan-out pattern?
How does boosting differ from bagging?
What is the fundamental difference between DBSCAN and K-Means?
If we have tools or MCP — how does LLM invoke them?
How are embedding models trained? How would you write your own embedder?
Have you worked with object-oriented programming?
How to extract text from scans?
What open source model do you use for local deployment via Ollama?
How would you automate chunking when dealing with 10 gigabytes of data?
How to evaluate LLM answers if they are correct but written differently?
Where is the prompt stored and how is it displayed in the monitoring system?
How to evaluate the quality of search in a vector database without annotations?
How to build a quality question and answer pool for system evaluation?
How did you implement Graph RAG? Why did you use Qdrant if it doesn't natively support graphs?
How would you write retry logic when calling an LLM? Are you familiar with the Tenacity library?
What is the token generation speed (tokens per second) in the Qwen model?