House of ZK sat down in Singapore with Rolf Versluis - Co-founder of Horizen and growth lead at zkVerify - right after our Verifying Intelligence program during Token2049. The happy hour buzzed behind us while we talked origins, zero-knowledge proof verification, and why business-grade privacy must be practical, fast, and compliant.
Alice - Rolf, thanks for making time on a busy day. For readers meeting you for the first time, how did you find your way into Web3 and ZK, and what led to co-founding Horizen?
Rolf - My path is a little unconventional. I started as an electrical and nuclear engineer, served as a submarine officer, then moved into semiconductors and data networking at Cisco before building and exiting a networking company. In 2015 I jumped into Bitcoin mining, then GPU mining. When Zcash arrived, private payments immediately clicked for me as a business need - you don’t want competitors seeing your customers or payroll. With Rob Viglione, we launched what became Horizen with sustainability in mind - treasury funding to a non-profit foundation so the protocol could keep improving. The thesis was simple then and now: permissionless, secure payments need credible privacy and an upgrade path.
Alice - Horizen built with privacy at the core. How did that evolve into focusing on zkVerify - a dedicated proof verification layer?
Rolf - Zero knowledge started for us with private transactions, but the bigger insight is broader - ZK lets you prove a computation happened correctly without revealing the underlying data. Generating proofs is only half the story. You also need to verify them, reliably and cheaply. Ethereum added precompiles that make verifying a specific proof system - Groth16 - inexpensive. That unlocked ZK rollups, which was huge. But the field kept innovating - Halo2, PLONK variants, STARKs, zkVMs - and we saw teams wrap modern proofs back into Groth16 just to verify on EVM. That adds latency and cost. zkVerify was our answer: a purpose-built L1 that natively verifies many proof systems, so builders can use the right prover and verify without expensive, slow wrapping.
Alice - What does the architecture look like, and why a layer one?
Rolf - We built on Substrate for its battle-tested components - nominated proof of stake, governance, pallets - and added pre-verification pallets for multiple proof systems. That gives us cost control, fast finality, and an upgradable path as ZK evolves without hard forking every time a scheme changes. Practically, a prover submits - say - a Halo2, UltraPlonk, UltraHonk, or STARK proof directly to zkVerify. Our validators check it and finalize on-chain. For EVM apps, we batch receipts and post them to a verifier contract so on-chain logic can trust a specific proof was verified on zkVerify. It’s a security spectrum - if you absolutely need native on-chain verification on Ethereum, fine. But for a large class of apps, a fast, low-cost, decentralized verification layer is the unlock.
Alice - Let’s talk numbers. How do time and cost compare with verifying on Ethereum?
Rolf - If you wrap a STARK in Groth16 and verify on Ethereum, you might wait 30 to 60 seconds and pay 5 to 20 dollars in gas. For high-value settlements that’s acceptable. For games, private polling, or interactive apps, it’s a non-starter. On zkVerify we target sub-10 seconds and sub-cent costs. That changes what’s feasible.
Alice - The ecosystem has many provers and DSLs. How do you handle the fragmentation?
Rolf - By embracing it. We support multiple schemes and welcome community verifiers. Distributed Lab contributed a Plonky2 verifier - plonky2 is great for sub-second mobile proving. We prioritize verifiers based on real usage so the chain is economically healthy - gas in ZKV pays validators, stakers, and the protocol. But we’re open source and happy to merge high-quality contributions, because the long-term win is being the neutral hub where proofs get verified.
Alice - You’ve been running hackathons and a long testnet. What did you learn that shaped mainnet?
Rolf - Documentation matters - a lot. We invested in examples, better SDKs, and DevRel so teams can go from “hello proof” to production. We simplified the Web3 integration path - one wallet experience even if you’re paying zkVerify fees and interacting with an EVM app. And we learned where builders live - university blockchain clubs and fast-moving dev communities. That’s why we pair hackathons with incubation and then grants so promising ideas get to market, not just to a demo.
Alice - Give us your favorite real-world use cases that benefit from scalable verification.
Rolf - Two categories stand out.
First - selective disclosure identity. Our partners at Self let users scan the NFC chip in a passport and derive verifiable credentials. You can prove “over 18”, or “from this province”, or “KYC-verified” - without dumping your entire identity everywhere. That’s perfect for compliance-aligned privacy in fintech, age-gated services, token sales, voting, and more.
Second - confidential compute plus ZK. Teams like Phala run workloads in TEEs so data and models stay secret. Pair that with a ZK proof that the code ran correctly and you get verifiable results from a black-box environment - think private DEX logic, confidential auctions, enterprise analytics, or AI inference on proprietary models. zkVerify is the neutral place to attest those proofs quickly and cheaply.
Alice - You mentioned games and interactivity. I heard you even shipped one.
Rolf - Guilty. I wrote a simple lemonade-stand game that generates a proof your score followed the rules over seven days. Teammates followed with blackjack and slot machine demos that prove fair play. They’re playful, but they teach the core idea - you can prove “this process ran by the rules” without revealing everything inside it. That mental model carries straight into enterprise and DeFi.
Alice - Where does Horizen fit now, and how does it relate to zkVerify?
Rolf - Separate missions, complementary tools. zkVerify is a neutral L1 for verifiable computation serving Web2 and Web3 - proofs in, receipts out, broad ZK coverage. Horizen is relaunching as a Layer 3 app chain on Base focused on business-grade privacy that’s compatible with regulation - think selective disclosure, viewing keys, and TEE-assisted workflows. Moving from a bespoke L1 to Base lowers cost, taps a thriving ecosystem, and lets us ship faster. There’s a clean bridge between the two - Horizen apps can rely on zkVerify receipts when they need third-party proof verification.
Alice - Privacy, but practical. How do you frame the compliance piece?
Rolf - Most businesses aren’t trying to disappear - they want competitor-proof operations and the ability to disclose under lawful process. That means privacy by default for day-to-day, with selective disclosure when required. ZK plus TEEs and good key management makes that feasible. If we want mainstream adoption, our rails must be better than today’s - faster, cheaper, and designed so businesses can meet obligations without oversharing.
Alice - For developers ready to build, what are the next steps?
Rolf - It’s permissionless - start today.
Alice - Perfect. Congrats on mainnet, and thanks for the candid download.
Rolf - Thanks, Alice. Great to be here - and great to keep building together.
