Alice: Shyam, Emanuele, thank you both for joining me today. I’m really excited to sit down with you and dive into the work you’re doing with Provably. Let’s start with your background - how did you both get into the Web3 and ZK space?
Emanuele: Thanks for having us, Alice. My background is in mathematics. I went through the usual academic path and then spent years in industrial research and development, primarily working on algorithms across machine learning, cryptography, and distributed systems. Around 2016-2017, I got into building prediction markets and working with digital assets. That’s when I really started focusing on cryptography, especially zero-knowledge proofs.
Shyam: My journey started a bit differently. I first discovered crypto in 2015 and, like many, fell down the rabbit hole - reading a ton of papers and diving into the ICO boom. But professionally, my entry into this space really began in 2018. I was leading a program at IBM in the UK focused on building connected and electric vehicles. Through that, I saw first-hand how smart products were generating a tremendous amount of data, but there was no real infrastructure to collaborate over that data securely. That realization led me to explore cryptographic solutions, and that’s how I met Emanuele. We both saw the potential of ZK and knew we had to build something ourselves.
Alice: That’s a great progression. You both come from different backgrounds, but you found common ground in the challenges surrounding data collaboration. That leads perfectly into my next question - what inspired the vision behind Provably? Can you expand on your idea of ‘collective intelligence’ and how Provably fits into the broader ZK landscape?
Emanuele: Absolutely. I like to say, “It’s the data, stupid”, paraphrasing an old political slogan. No matter how advanced our algorithms were, the biggest challenge was always data access. We’d spend a lot of time trying to convince partners to share their data, and despite legal contracts and agreements, it just wasn’t happening.
The core problem was clear - people don’t want to share data. So, we asked ourselves: how do we enable insights from private data without requiring it to be shared? That’s where Provably comes in. We provide a way to generate verifiable analytics on private data, ensuring data privacy while still allowing collaboration and monetization. We’re not just helping enterprises work together but also laying the foundation for broader, decentralized ecosystems.
Shyam: Right. We used to operate in silos, with data hoarded by individual entities. But the world is becoming increasingly networked, and businesses want to collaborate. The challenge is how to do that efficiently, securely, and at scale. That’s what we set out to solve with Provably.
Alice: That makes sense. You’re tackling the core issue of data privacy while still enabling data-driven collaboration. Given that, can you talk about the status quo of data management today? What are the gaps, and what challenges exist with current solutions?
Emanuele: Today, data sharing is largely built on legal contracts, trust, and access control mechanisms. A company masks some data, grants access to a partner, and they operate under strict agreements. But in practice, the process is slow, inefficient, and still involves a level of trust that is hard to verify.
There are also some advanced cryptographic approaches, like fully homomorphic encryption, but they are often too slow or impractical to scale. The reality is that data collaboration is still clunky and inefficient.
Shyam: Exactly. Current solutions don’t scale well for real-time collaboration. You need a way for data to remain private while allowing multiple parties to extract insights securely. That’s where zero-knowledge proofs come in. With Provably, you don’t need to share raw data - you just generate ZK proofs over the data, ensuring privacy while still enabling computation.
Alice: That’s a great approach. You also touched on another key challenge - adoption. Many developers and businesses aren’t familiar with cryptography or Web3. How does Provably address that barrier?
Shyam: We focused heavily on accessibility. Provably is delivered in a containerized format, deployable in any cloud or on-prem environment, and interacts with data through simple APIs. Most importantly, we built it to work with SQL. If you can write SQL, you can run verifiable analytics with Provably.
SQL is one of the most widely used languages in data science, with around 7 million developers familiar with it. By abstracting the complexity of ZK proofs, we make it possible for businesses to leverage this technology without needing to learn new paradigms.
Alice: That’s impressive - reducing the learning curve is crucial for adoption. On that note, can you walk me through a typical user journey with Provably?
Shyam: Sure. Mid-October, we’re launching Provably 1.3, which is designed for two main user groups: data owners and insight seekers.
The key innovation is that all of this happens while maintaining privacy. No raw data is shared - only proofs and aggregated insights.
Alice: That’s a game-changer. Now, beyond blockchain applications, what other industries could benefit from Provably?
Emanuele: Healthcare is a big one. Regulations around patient data are strict, making cross-border data collaboration very difficult. With Provably, hospitals and researchers can analyze sensitive data while ensuring compliance and privacy.
Another example is content rights and royalties. Platforms like YouTube, Spotify, and Netflix calculate payouts to artists and creators based on opaque, internal metrics. With ZK proofs, they could generate verifiable statements proving exactly how royalties were calculated - without exposing user data.
Alice: That’s a fascinating use case - bringing transparency while preserving confidentiality. Shifting gears a bit, I’d love to hear more about your technical design choices. You’re using Bulletproofs and Pedersen commitments - why did you choose those?
Emanuele: Our design principles were clear: we needed something fast, secure, and well-tested. Bulletproofs fit well because they are non-trusted setups, meaning we don’t require a ceremony. They also support range proofs and inner product arguments, which are useful for analytics.
Pedersen commitments offer excellent privacy guarantees without relying on trusted setups, making them ideal for our use case. We opted for battle - tested cryptographic tools rather than experimental techniques, ensuring security and performance.
Alice: That makes a lot of sense. Before we wrap up, any final thoughts or advice for those looking to build in this space?
Emanuele: We’re just at the beginning of unlocking the full potential of data collaboration. The convergence of crypto, AI, and privacy technologies is happening, and we’re excited to be at the forefront of it. There’s so much opportunity for innovation in how data is used and monetized.
Shyam: I’d add that ZK is now mature enough to extend beyond blockchain applications. For a long time, we were limited to on-chain use cases, but with ZK analytics, we can finally explore trustless collaboration over off-chain data. If you’re an entrepreneur, engineer, or researcher, now is the perfect time to build.
Alice: That’s an inspiring note to end on. Shyam, Emanuele, thank you both for such an insightful conversation. I’m looking forward to seeing where Provably goes next!
Shyam & Emanuele: Thanks for having us!
Follow @ProvablyShyam, @EmanueleRagnoli, and @GetProvably on X for more insights.
