Банк вопросов из реальных собеседований

How to build a service that generates a presentation and PDF by text query?

Design a multimodal fashion compatibility recommender: candidate generation, embeddings, outfit labels, triplet mining, reranking and typical failure modes.

There are two identical heterogeneous ropes. Each burns completely in 60 minutes, but the burning rate along the length is unknown. There is a lighter. How to measure 45 minutes?

The user uploads a photo of the product. When is it enough to select an object with a limiting frame, and when is it necessary to search for an exact segmentation?

The user again sent a request to replenish the balance. How to make endpoint idempotent?

How does Redis process commands, and why are individual commands usually atomic from a client perspective?

How to adapt the existing model to new tasks? What options would you consider and what does the choice depend on?

The agent generates presentations/tables/documents. How to store versions of model files and support rollback/editing?

The dataset has missing values and noisy features. How to systematically process them before training and in production?

A legacy ML platform has accumulated technical debt. How do you decide whether to rewrite it or keep improving it incrementally?

In a microservice system, services communicate through API and events. How to document and check contracts so that releases do not break consumers?

How to justify the choice of CatBoost for tabular time series features and what simple solutions to compare it with?

In PyTorch, what should Dataset do, what should collate_fn do, how do num_workers affect this, and where should .to(device) usually happen?

Multiple threads update individual pixels of the same screen. What can go wrong, and how would you design synchronization?

You pull out visually similar fashion products, but there are a lot of almost identical things on the final list. How do you rank the list to maintain relevance and add variety?

How would you split data for validating a fraud model? What leakage risks would you check?

Review a simple PyTorch multiclass image-classification training loop. What bugs and design issues would you look for?

What are the risks of applying reinforcement learning to trading and maintaining market liquidity?

Multiple clients send pixel updates to a central server over the internet. What transport/protocol would you use and what tradeoffs matter?

What techniques can compress or adapt a large model, and how would you detect and reduce catastrophic forgetting during fine-tuning?

For realtime CTR dashboard, you need to describe Kafka/event log. What event schema is needed and by what key to partition?

How should an agent-generated table be stored so users can sort, filter, and export it without asking the model to regenerate the result?

The user created a presentation, a PDF and a spreadsheet, and then asked to “sort it out.” How do you know what the team is about?

How to test and roll out prompt changes in a LLM product?