Description
You need to prove you have the money without showing the balance, prove the computation ran without running it again, prove you belong on the list without handing over the list. It sounds like magic, and that is the trap: every explanation either waves its hands and calls it magic, or drops you straight into polynomial commitments with no bridge in between. So you can say zero-knowledge proof in a meeting, but you cannot tell which problems it actually solves, which it does not, what a proof really guarantees, or what each one costs to verify on chain. This is the frontier skill the best get paid the most for, and it stays locked behind math nobody bothered to translate. What you need is the working model, not a cryptography degree.
This book builds the working model of zero-knowledge proofs an engineer needs to use them, not the full cryptography curriculum. It starts from what a proof actually is, the three properties that matter (completeness, soundness, zero-knowledge), and walks the bridge most explanations skip: from an interactive proof where a prover and verifier talk, to a non-interactive one you can post on chain, via the Fiat-Shamir trick. From there it builds the pieces you actually touch: how a computation becomes an arithmetic circuit and a constraint system, the honest tradeoffs between SNARKs and STARKs (proof size, verification cost, trusted setup, post-quantum), what the trusted-setup ceremony is and what breaks if it was rigged, and how commitments and Merkle proofs let you prove membership without revealing the set. Then it gets concrete: writing a small circuit, verifying a proof on chain and what that costs in gas, and the real applications, ZK-rollups that scale a blockchain by proving thousands of transactions at once, and privacy systems for payments and identity. The goal is judgment: knowing when a zero-knowledge proof is the right tool, what it guarantees, and what you pay for it. For engineers who want to understand ZK well enough to build with it and to call out the hype.
Who this was written for
This book is for: engineers who keep hearing that zero-knowledge proofs let you prove something without revealing it, and want to actually understand how that is possible and how to use it, without a cryptography degree.
What you’ll walk away with
- Chapter 1: The bank that asked for a number you swore to hide
- Chapter 2: Drop one of these three and your proof protects no one
- Chapter 3: Why your proof breaks the moment no one is left to answer
- Chapter 4: The only thing a zero-knowledge system can actually prove
- Chapter 5: Why there is no best proving system
- Chapter 6: The 2 KB blob that can quietly drain the whole contract
- Chapter 7: Write a circuit like code and you ship a hole
- Chapter 8: Showing you belong to a set without revealing the set
- Chapter 9: Getting a contract to say yes without learning a thing
- Chapter 10: When a $10 transfer costs more than $10 to settle
- Chapter 11: Paying for something and leaving no trace behind
- Chapter 12: Two features that look like ZK, and only one is
- Chapter 13: Becoming the engineer who knows when not to use ZK


