Python
In which case is the principle of open-closed violated?
How to create a data structure that cannot have values inserted or modified but can be accessed by key?
The description is provided by a business analyst who wrote the task "as they could" and can give business requirements descriptions. There is a terrible legacy PHP code that receives log files from S3 parses them, writes the result to the required location. It is known: files arrive once an hour, currently 5 files all files are in one bucket Data should be directed to different places depending on the format – postgres/local file It is necessary to write an ETL that implements this functionality. Writing specific implementations of methods (reading data/sending data, etc.) can be neglected.
What parts does JWT consist of?
Tell about yourself: experience, what you like, what you are looking for?
How do you find the described scope of work and tasks?
How do we determine the required infrastructure capacity for a model?
Two tables with 100 rows each with arbitrary numbers. What is the minimum and maximum number of rows that FULL OUTER JOIN can return?
What is Mock and Spy in testing?
What good things did you do at [company]? Tell us about interesting projects.
If the quality of the team’s work deteriorates, tasks take longer, and they do not fit into story points — how to understand the reason?
When should you use asynchrony, and when should you use synchrony/multithreading?
Tell us how deeply you understood how Python works under the hood.
How do Kafka and RabbitMQ fundamentally differ?
What is the difference between a RabbitMQ consumer and a Kafka consumer?
What is CQRS and how have you applied it in a project?
Regarding relational databases in general (not specifically about PostgreSQL): if the database cannot handle the volume of data, and read/write operations are slow — explain step-by-step, from simple to complex, how to optimize this situation.
Have clear specifications been provided to you, or have you participated in discussing requirements and communicating with analysts, PMs, and the business?
What will you do if a heavy SQL query becomes a bottleneck?
What are p95, p99, and what other percentiles are there?