Machine Learning / AI
Why does accuracy show an inadequate quality measure when there is class imbalance?
What loss function did you use when fine-tuning the embedder?
What was actually more difficult than initially expected?
Did you come to the RAG project from the very beginning or at some stage?
What are items? How is it correctly pronounced?
What does an 'interesting task' mean to you at work?
Please tell how dictionaries work under the hood in Python, and what about collisions?
What is tokenization?
How did you implement Text-to-SQL? How did you measure 91% accuracy?
What is a streaming API?
Which companies are you currently at some stage with?
-- # -- Marketers launch promo campaigns in the app. There are two tables: -- # -- campaigns - list of campaigns -- # -- - campaign - campaign name -- # -- - action_type - campaign type: "push" or "banner" -- # -- communications - backend log of sending communications of these campaigns to users -- # -- - user_id - user identifier -- # -- - campaign - campaign name -- # -- - status - event status: "success" or "error" -- ------------------------------------------------------------------------------------------------------------------------ -- # -- 1. Initially, the campaigns table has 4 rows: -- # -- | campaign | action_type | -- # -- |-------------|-------------| -- # -- | promo_dogs | push | -- # -- | promo_cats | banner | <- randomly duplicated -- # -- | promo_cats | banner | <- two rows -- # -- | promo_rats | push | # # -- Also known that: -- # -- campaigns promo_dogs and promo_cats were successfully conducted on 100 users each, and each user received one communication, -- # -- while promo_rats is only planned. -- # -- 1.1 What will the query output: SELECT campaign, COUNT(*) as rows, COUNT(DISTINCT co.user_id) AS users FROM campaigns AS ca INNER JOIN communications AS co ON ca.campaign = co.campaign GROUP BY ca.campaign AS campaign -- # -- 1.2 How will the answer change if you change the JOIN type to LEFT? -- # -- 1.3 List the order of execution of the operators in this query. -- Answer: promo_dogs ... ...
Why is the analytical formula for linear regression not used in practice?
Tell me about the geometric meaning of regularization.
How to write the condition == False more readably in Python?
Can we pass a list as an argument to a function and then modify it inside? What will happen to the old list?
Can you recommend a second candidate for the position?
Tell us how the cooperation with a foreign company on an individual entrepreneur basis is arranged — with a legal entity in Russia or how it is built?
What other metrics should be added for monitoring document updates (to avoid discovering that after an update, only 3 documents remain instead of 10,000)?
How would you now (with current knowledge) solve the classification and routing of requests in a service desk, using a multi-level approach rule-based → ML → LLM fallback → human? Name specific technologies, models, and metrics.