Банк вопросов из реальных собеседований

Explain the difference between a Kubernetes pod, service, deployment and node.

Explain the difference between a Kubernetes pod, service, deployment and node.

What is data drift, how would you detect it in production, and what can you do when it starts hurting an ML model?

Какие логи, метрики и алерты нужны после запуска рекомендательной модели?

Tell us what classes of models are in recommendation systems and where they are usually used.

You have a multi-GPU server and want to host one or more open-source LLMs. What software stack and design choices would you use?

The basket changes in the current session. How should the recommendation service take into account the addition and removal of goods, disable the cache and choose a backup response?

How do you estimate the amount of data required to recognize audio events and decide when a device should enable a more expensive processing mode?

What does it mean to take an ML model from training to production, and which pieces should an ML engineer be able to own?

Where are the main costs and errors in creating rare scenario simulations and how to use them safely in a search engine model?

How do online and offline recommender designs differ when model latency, feature cost and freshness matter?

How to organize an online heavy recommendation model and meet a set delay budget?

Shipping cost models need a fresh sign of courier load. What events and states do you count on to use at the time of making a decision?

A neural network inference pipeline is too slow. What optimizations would you consider before changing the model architecture?

How to consistently optimize the routing, packing, latency, bandwidth and service cost of a language model?

The call-processing pipeline works but inference is too expensive. What would you optimize first?

What main architecture families are used for generative models, and where are they commonly applied?

What should a ML engineer do to bring the model to a production service: interface, model file, Docker, monitoring and updates?

Your punctuation model has good validation metrics, but users complain that some outputs are strange. What could be happening and how would you debug it?

What are the differences between recall@K, precision@K, coverage and NDCG in evaluating recommendations? How do they behave when they change K?

What are the differences between recall@K, precision@K, coverage and NDCG in evaluating recommendations? How do they behave when they change K?

What are the differences between recall@K, precision@K, coverage and NDCG in evaluating recommendations? How do they behave when they change K?

How would you evaluate the full search pipeline and its individual components offline and online?

What are the differences between LLM initial input processing and step-by-step generation, and why do they need different optimization methods?