Банк вопросов из реальных собеседований

The command changes prompts/models/rules for invoice parsing. How not to break quality with every change?

What problems arise when using long context in LLM and what approaches do they address?

How do MQA, GQA, and MLA reduce attention during generation compared to conventional multi-headed attention?

How does a sparse MoE model differ from a dense one, how does a router choose experts, and what system difficulties arise in application?

The ranking model is ready. How to put it into production: offline batch or online inference?

We need to build a support bot for a fintech application. What components are needed and how can you reduce the risk of answering incorrectly?

Как бы ты сделал retrieval-augmented generation для короткого factual snippet в поисковой выдаче?

How to build a RAG/search system if the case is similar in scale to a large web search?

How do you design an assistant who answers the current lesson but doesn’t disclose future material?

How to transfer Vision Transformer from individual images to a time sequence without making archive indexing too expensive?

You need to design a product: the user gives a text problem, the system makes a presentation with slides, tables and pictures. How to build a pipeline?

Нужно спроектировать AI-native продукт, который по запросу пользователя генерирует качественные motion graphics. Как выбрать между pipeline и fully agentic архитектурой, как встроить human-in-the-loop evaluation и как управлять trade-off между quality, consistency и latency?

How to select a time window and combine frame views into one or more vectors for search?

There are several GPU workers in LLM serving. Why naive round-robin routing may be ineffective, and how to design an adaptive routing layer that takes into account GPU headroom and KV-cache reuse?

We need to build a system that extracts useful fields from PDF invoices from different suppliers. Which architecture should you choose?

How to check online a new model for selecting security questions if an error could miss a fraudster or block a client?

How would you build item embeddings from text, images/video and categorical/numerical attributes under real serving constraints?

After gradient boosting, which neural network architecture can be tested for sequential data and on what properties of the task depends the choice?

In international search, you need to show a short description of the object in the response card, for example for China. How to build a ML system for generating such descriptions?

Design a semantic search layer for geo/address suggestions where users can type categories like "cafe" and expect restaurants, POIs and relevant addresses across many languages.

What Unit Index, Update, Filters, and Evaluation Decisions Should Be Completed Before Adding a Re-Ranker?

How to build a dataset for a model that evaluates a user’s response to a shipping cost or minimum order amount?

How to independently code text and road segment, build a vector index and train a quick search for candidates?

How to design candidate generation and ranking for selecting suppliers for a procurement application in the corporate procurement marketplace?