Start here in Interview Questions
The most useful guides in this topic, not just the most recent.
Interview Questions Guides & Tutorials
Curated interview question banks with answers — SQL, Python, data science, Spark, Power BI, AWS, ADF and more.
Interview Questions articles
Kafka Interview Questions Explained
Kafka interviews test whether you can predict what happens to a message: which partition it lands in, in what order it is read, and whether it can be processed twice. Five questions are answered on a real Kafka 4 broker fed with an hour of real New York taxi pickups, including a crash that reprocesses 50 messages.
Generative AI and LLM Interview Questions
Many generative AI and LLM interviews now include questions you can only answer with numbers: how good is this retrieval, how many tokens will this cost, which search method should we use. This guide works through those questions on a real banking-queries dataset, with recall, MRR and token budgets actually computed.
MLOps Interview Questions and Answers
MLOps interviews test whether you can keep a model useful after deployment. This guide maps the rounds with model answers, then works a real case: a fare model trained on January 2024 taxi trips got worse in January 2025, while every drift metric stayed calm and retraining barely helped.
Git Interview Questions Every Developer Should Know
Git questions in interviews check whether you can undo mistakes safely, not whether you know every command. Six common questions are answered by running Git on a throwaway repository: merge versus rebase, the three reset modes, revert, recovering a lost commit, why .gitignore ignored nothing, and bisect.
Data Modeling Interview Questions Explained
Data modelling interviews test design decisions, and real data forces them. Seven questions are answered by building a star schema from a year of real online retail orders: declaring the grain, where price belongs, messy product names, customers who changed country, unknown members, a date dimension and additivity.
ETL Interview Questions for Data Engineers
ETL interviews test what happens when real extracts misbehave. Six questions are answered on the two yearly files of a real online retailer, which overlap by nine days: what a plain append does, how a watermark should compare, why there is no key, which quality rules need business input, and how to prove a load.