Five companies. One through-line.

People ask me how a career that started with a hardware startup ended up in tokenized stocks and onchain trading. On paper it looks like five unrelated jobs. A hardware product, a blockchain wallet and explorer, a privacy sidechain, a decentralized exchange, and a trading app built on Solana. From the inside it has always felt like the same job, done in different rooms.

This is the longer version of that answer. I'll go newest first, because that's who I am now, and end with the chapter that taught me the most expensive lesson.

Lume: the money path is the product

I co-founded Lume and ran it as CEO from April 2025 until the chapter closed for regulatory reasons in July 2026. It was a decentralized app for buying tokenized U.S. stocks with USDC. Think of something like Robinhood, but onchain and self-custodial, for people who couldn't easily open a U.S. brokerage account. We added perpetuals and an AI trading co-pilot on top.

In four months we processed more than $4.1M in trading volume across 4,100+ users, and shipped 78+ production releases in five months across web, iOS, Android and Solana Seeker. The number I'm proudest of is less flashy. Time from onboarding to a first funded trade went from weeks to under 60 seconds.

That number is the whole lesson of the chapter. Nobody cared that the stocks were tokens or that settlement happened on Solana. They cared whether money got in, whether a trade went through, and whether they could get it back out. I wrote about that funnel in From weeks to under 60 seconds, and about how we kept the referral program from paying farmers in When referral farming breaks your growth loop.

The hindsight is just as useful. We concentrated in one country because that's where early users showed up, and when that country squeezed its crypto on-ramps, our backups failed together. If I did it again I'd spread across regions earlier, even if each one started smaller. I wrote about that in India keeps squeezing the on-ramp and about closing a company well in Shipping when regulation closes the chapter.

Sonic: a launch is a product too

Before Lume, I co-founded Sonic on the Internet Computer Protocol and ran it as CEO from September 2023 to January 2025. With a five-person team we built an automated market maker, a DeFi hub and a Liquidity Bootstrapping Pool. The exchange crossed $200M in cumulative volume, with a $13.4M peak month. Projects in the ICP ecosystem raised more than $9M through our LBP launches. We raised $6M for Sonic itself through a community DAO sale, and the community voted to create Sonic DAO.

Sonic taught me that a token launch is not a marketing event with a contract attached. It's a product with users on both sides. The project needs mechanics it understands, a launch plan and liquidity afterward. The buyer needs a fair price and a reason to stay. When I treated the launch flow, the advisory work and the community as one product, launches went better. I went deeper on this in An LBP is an access product.

IOHK and Midnight: privacy has to live in the wallet

From January 2022 to August 2023 I was a Senior Product Owner at Input Output (IOHK), working on Midnight, Cardano's data-protection sidechain. I owned four product areas: the Midnight wallet, a private DAO, the Pub-Sub Indexer and the blockchain stack underneath them.

Owning the top and the bottom of the stack at the same time was the education. Zero-knowledge proofs are remarkable, but a user never sees a proof. They see a wallet, a confirmation screen and a vote. If privacy only exists in the protocol and the wallet treats it as an afterthought, people won't trust it or won't use it. The interface is where the guarantee becomes real. That's the argument in Privacy products fail when the wallet is an afterthought.

I also did the Carnegie Mellon Tepper product management program during this period. It didn't change what I believed so much as give me better language and tools for it, which mattered when I had to explain tradeoffs to engineers, researchers and business stakeholders in the same meeting.

NEM: the explorer and the academy are product

From 2018 to 2021 I was a Product Strategist at NEM Group, working across the Catapult Mobile Wallet, the Catapult Block Explorer and NEM Academy. Three launches, and for long stretches two product backlogs running at the same time.

What stayed with me from NEM is that the "supporting" products weren't supporting at all. A block explorer is how people check that what the wallet told them is true. An academy is how a new user learns enough to try the wallet in the first place. I spent a lot of time as the bridge between business stakeholders and developer teams, writing launch materials and tracking product performance, and that taught me that a roadmap isn't finished until someone outside the building can understand what shipped and why.

Exploride: de-risk everything you can

My first company was Exploride, which I co-founded and ran as product owner from 2014 to 2017. It was a hardware product, funded through Indiegogo and private capital, and it took years to build. It didn't work.

I don't lead with it, but I'd be lying if I said it didn't shape everything after. Hardware punishes every assumption you didn't test early, because each change costs months. The lesson I carried out is one I've heard Jeff Bezos make, and I agree with it completely: a startup is already risky, so remove every risk you can. Test the demand before you build the factory. Make the first version smaller. Find out what breaks while it's still cheap to fix.

The through-line

Lined up, the five chapters say the same thing. The technology was never the hard part. Hardware, Catapult, zero-knowledge proofs, AMMs and tokenized equities all worked well enough. What decided whether anyone used them was the thin layer where a real person had to trust the thing: the first funded trade, the wallet confirmation, the explorer that checks the wallet, the launch that has to be fair. Even Exploride fits, as a product that took years to build before the market could tell us whether it was right.

That's what I mean when I say I build products that make complex technology useful. It isn't a slogan about simplicity. It's a choice about where to spend product attention. I'd rather own the onboarding, the money path and the launch than another feature nobody reaches.

• Find the moment of trust in your product and measure it directly, the way we measured time to first funded trade.
• Treat the supporting surfaces, like explorers, docs and launch flows, as real products with owners.
• Put the hardest guarantee in the interface, not only in the protocol.
• Spread your dependencies, whether that's regions, on-ramps or suppliers, before you need to.
• De-risk early and cheaply, because the market will do it for you later and expensively.

What I'm optimizing for next

I'm looking for my next product leadership role, on a team putting real money on new rails: stablecoin payments, tokenized equities, onchain trading and the products that make them safe for ordinary people. Those are the problems where the technology is ready and the trust layer still isn't, which is exactly where I've spent the last twelve years. If that's what you're building, I'd like to hear about it at sunil.vallath@icloud.com.

The structured version of this story lives on About, current focus is on Now, and the case studies are on sunilvallath.com. The short answer is Who is Sunil Vallath?, and everything else is on the writing hub.