Credi is an on-chain private credit platform: a stablecoin vault on Ethereum. Financing short-duration VISA Ramp card receivables. Behind it is a founder who spent over a decade underwriting credit before writing a single line of it on-chain. We sat down with Alessio Marinelli, CEO of Credi, to talk about credit risk, real yield, and why boring is a feature.

What inspired you to create Credi?

DeFi solved distribution. Anyone, anywhere, can allocate capital into a vault in minutes. What it hasn't consistently solved is what that capital is actually earning. Too much of the yield on offer is circular: ▫️Token incentives ▫️Funding rates ▫️ Leverage on leverage. When the music stops, so does the yield.

Meanwhile, in the real economy, there's an enormous, unglamorous machine that has generated returns for decades. Every card transaction has a settlement gap, and financing that gap is one of the oldest, most predictable credit businesses in existence. Credi exists to connect those two worlds, on-chain capital and payment receivables, without diluting the risk discipline that business demands.

Can you share a bit about your background and how it led to Credi?

I've spent my career on one question: how do you price and manage credit risk? That took me through more than fifteen years at KPMG and Deloitte, including leading Data Analytics for KPMG in Hong Kong and China, over a decade in structured finance across Asia and the UK, and a PhD in machine learning applied to credit risk.

Then I moved on-chain. I co-founded Monetalis and served as its Chief Risk Officer, where we managed MakerDAO's Clydesdale vault: $1.7 billion in real-world assets, the largest single RWA vault in DeFi at the time.

That experience taught me that on-chain credit works when the structure does the heavy lifting, not the marketing. Credi is the distillation of that lesson.

For readers who are new to Credi, how does it actually work?

Credi finances pre-approved card transactions inside the VISA Ramp ecosystem. We advance up to 80% of a transaction's value, and the settlement cycle runs 2 to 5 days. Short duration, high turnover.

The structural detail that matters most

Settlement flows directly from the regulated Merchant Acquirer into the vault. It bypasses the borrower entirely. That single design choice removes the classic repayment risk of lending. We're not waiting on a borrower to pay us back; the payment rails themselves deliver the funds.

Depositors access this through an audited Enzyme Permissioned Vault on Ethereum mainnet, with a three-month soft lock and a one-week cooldown. The return is a fixed rate targeting 15% net APY, rising toward 18% over twelve months.

A lot of DeFi products advertise double-digit yields. What makes Credi's different?

The source. Our yield comes from the economics of payment infrastructure, the fees merchants already pay to access their money faster. It doesn't come from token emissions, crypto funding rates, or anything that depends on market sentiment.

That's the distinction I'd encourage every allocator to interrogate, for any product.

→ Where does the yield actually come from, and does that source exist in a bear market?

Payment volumes don't care whether crypto is up or down. People keep buying things.

To date, we've recorded zero realized losses across the assets we've financed. We are careful with that number. Past performance never guarantees the future, and anyone who tells you otherwise shouldn't be managing your capital. But it reflects the structure working as designed.

What values and principles drive you as a founder?

Discipline over narrative. In credit, the moment you start believing your own story more than your data, you're already losing money. You just don't know it yet. I'd rather under-promise on yield and over-deliver on structure than the reverse.

The second one is transparency about risk. Every credit product has risk. The honest ones tell you where it lives and how it's mitigated. The dishonest ones tell you it doesn't exist.

What do you hope to achieve with Credi in the long run?

I want Credi to prove that on-chain private credit can be genuinely institutional.

Not "institutional" as a marketing word, but in the sense that a pension fund's credit committee could look at the structure and recognize the discipline. Payment receivables are our starting point because the duration is short and the settlement path is clean. But the machine we're building, with regulated settlement flows, structural borrower bypass, and transparent on-chain vaults, extends well beyond one asset class.

How do you see the future of on-chain private credit, and where does Credi fit?

The RWA sector has moved a lot, but most of the volume so far has been in one product: tokenized treasuries. That was the easy first step. Take the world's most liquid asset and put a wrapper on it.

But Private credit is harder. It's a multi-trillion-dollar market in traditional finance, and it's precisely where on-chain rails add the most value: transparency into collateral, programmable settlement, global capital access. But it demands real underwriting. You can't fork a credit desk.

That's where Credi is positioned to lead.

→ We didn't add credit expertise to a crypto team, we brought crypto to a credit team.

What's the most important lesson you've learned in your career?

That risk you can't see is the only risk that hurts you. Every blow-up I've studied, in banking, in DeFi, anywhere, traces back to an exposure someone chose not to look at. So I've built my career, and now Credi, around a simple habit: go looking for the uncomfortable number first. If the structure survives that scrutiny, everything else is execution.

-- Credi's $USDC vault is live on Turtle. Learn more at credilabs.io or follow @credilabs.

This article is for informational purposes only. It is not investment advice or an offer to sell securities.