Blockchain Technical Interview Questions (2025 Edition)

2025-07-06

Compared to the 2023 version of Blockchain Technical Interview Questions, which focused more on a macro perspective, this edition leans slightly more toward engineering practice and includes more technical details. These two versions complement each other; they are not upgrades. The questions here are purely based on my personal experience, just like many interviewers who only ask what they themselves know and will not ask about things they do not understand. So the scope and depth of these questions are inevitably limited by my own level of knowledge:

  1. Why is the Ethereum client divided into execution layer and consensus layer?
  2. How does Ethereum’s PoS process work, and how do you initialize a PoS network?
  3. What are the soft fork and recovery mechanisms in Ethereum PoS? Is Cardano’s PoS the same as Ethereum’s?
  4. What types of Ethereum nodes exist, and what are their use cases?
  5. Why is EVM execution single-threaded? Why has no team in the world succeeded in building a “parallel EVM”?
  6. Does the Solidity language have garbage collection (GC)? How does it handle dynamic memory allocation?
  7. In what scenarios is inline assembly necessary in Solidity?
  8. Are you familiar with the PBFT consensus? What’s its general process?
  9. What is the formula for PBFT fault tolerance, and why that number specifically?
  10. Why does PBFT require a second round of voting?
  11. What is Solana’s consensus mechanism like? Is TowerBFT voting on blocks?
  12. Why can Solana’s smart contracts be executed in parallel while Ethereum’s cannot?
  13. What is the upgrade process for Cosmos nodes? How is it different from Ethereum’s? What are the risks of this model?
  14. What is the general flow of an Op Rollup? Which part does ZK Rollup optimize based on Op Rollup’s model?
  15. How does asset bridging work between Ethereum L2s? Technically, how does it differ from bridging assets across entirely different networks?
  16. Ethereum recently had a major upgrade introducing EIP-7702—what is it for? How is it related to Account Abstraction (AA) wallets?
  17. What interesting blockchain-related technical topics have you personally explored?

That’s all the questions I can think of at the moment. For more in-depth, engineering-oriented topics, I only have a general idea of the direction—I haven’t personally worked on them. Still, my experiences over the past couple of years have been relatively rich. Comparing the two versions of the question list reveals a lot of changes. I hope I can continue to improve and not lose my way.

If you’re a job seeker in the blockchain industry—especially if you’re a relatively inexperienced engineer—don’t be intimidated by the questions above. In real interviews, you’ll rarely encounter such deep and thought-provoking questions. More commonly asked questions include things like: “What are the common fields in an Ethereum transaction?”, “How do you cancel a transaction that’s already been sent?”, “What is a reentrancy attack in Solidity?”, “What components make up the Op Stack?”, “What is create2 in Ethereum contracts?”, etc. Go confidently into job hunting—there aren’t that many people who truly understand the technology deeply.

There’s currently an issue in the blockchain industry: there is no systematic theoretical framework, only fragmented, cutting-edge engineering experiments. Compare this with programming languages: from Church’s and Turing’s models of computation to functional programming languages, compilers, and type systems, decades of development in academia and industry have produced both highly abstract theoretical foundations and practical industrial applications. The field is relatively mature. Blockchain, however, is still new. It emerged in 2008 and began entering public awareness around 2013. That has not been enough time to establish a solid academic system. Instead, projects go their own ways, creating separate standards and defining their own terminology. Account models, consensus, contracts, and cross-chain systems all differ from one blockchain to another. Could anyone unify blockchain into a coherent theoretical framework? Even Vitalik could not do it; perhaps there would be hope if Satoshi returned.

That is why there is no need to put much faith in university “blockchain majors.” How could teachers and professors who have never left academia already have had enough time to abstract blockchain technology into theory, write textbooks, develop curricula, and teach it to students? Consider how long that cycle would take. For the same reason, do not place too much trust in published technical books. Publishing a book takes years, and by the time it appears, the world has already changed. A Bitcoin conference is scheduled for the second half of this year, where the team that invented the concept of inscriptions two years ago may announce something new. How could a university course or book keep up with that pace? The most advanced technology in the industry can only come from the concrete exploration and experimentation of individual projects, which naturally results in a field that lacks a unified system.