Practicing SQL queries under time constraints is a baseline exercise. Solve the following leetcodes to keep yourself sharp.

https://leetcode.com/problems/big-countries
https://leetcode.com/problems/swap-salary/
https://leetcode.com/problems/not-boring-movies/
https://leetcode.com/problems/duplicate-emails/
https://leetcode.com/problems/combine-two-tables/
https://leetcode.com/problems/employees-earning-more-than-their-managers/
https://leetcode.com/problems/customers-who-never-order/
https://leetcode.com/problems/delete-duplicate-emails
https://leetcode.com/problems/rising-temperature
https://leetcode.com/problems/classes-more-than-5-students
https://leetcode.com/problems/second-highest-salary/

League of Legends handles 1 million concurrent players, with over 100k concurrent 10-player games.

During games, players have different metrics like gold, equipment, score (kills, deaths, assists, known as KDA). There's an API that allows you to fetch this information in discrete requests, but we want to turn this into a real-time API to service analytics services.

An example product with real time data - https://probuildstats.com/champion/blitzcrank.

Build a high level design on how to turn an API into a real-time, scalable and reliable service.

Build a data aggregator for various blockchains and smart contracts - this can include different cuts and views of the same data, as well as different ways to read, transform and store information.

Build an architecture to service video over streaming services.