Why crypto needs its own banking infrastructure: The Fiat Republic origin story

The bank crypto kept asking for

I bought my first Bitcoin (0.5 BTC) in November 2017 at $7,000 per bitcoin, mostly because a friend wouldn't stop ribbing me about being a payments geek who'd never touched crypto. Then I sold it in December near the $20,000 high. In a hasty fashion common in our industry, I bought it back near the highs a few days later, only to subsequently watch it crash all the way back down again. Welcome to crypto.

We all have a crypto story like this. But what got me hooked wasn't the trade. It was the operations. By 2018 and 2019, I was running algorithmic arbitrage between Bitfinex, Bitstamp and Coinbase. The premise is simple: tiny price differences across exchanges, captured by moving fiat between them faster than the spread closes. The premise is also, practically speaking, nearly impossible, because the bottleneck isn't the exchange. It's the bank.

My bank flagged me constantly. Five identical transactions in a day were enough to repeatedly land me in the enhanced due diligence queue. I was also made to sign a waiver absolving them of any responsibility for my money once it touched a crypto venue. The exchanges' own banks were no better, frozen every few transactions. Compliance requests were endless. Passport again. Utility bill again. Source of funds, again. I was sending the same documents back and forth every other day.

The barrage of requests reached a crescendo when, in August 2019, Coinbase lost its UK banking partner. Overnight, I couldn't top up with faster payments. The only option was a 3% card fee in and a 3% card fee out. Months later, Coinbase had found another partner. However, a significant number of UK customers had already left during that interim.

That was the moment that stuck with me. If Brian Armstrong, a public figure, licensed, working with regulators, and a horde of compliance staff, could be cut off by one of the biggest banks in the world overnight, what hope did anyone smaller have?

The pattern from the other side

In the years that followed, I kept seeing the same picture, but this time from the other side of the table. Crypto businesses, serious ones with licenses, compliance teams, and real revenue, were being turned away by banks every couple of weeks. Tether. Circle. Coinbase. Kraken. Crypto.com. Gemini. Everyone. Begging for any sliver of SEPA or Faster Payments access.

The reason was always the same, and it wasn't malice. Banks aren't built to take on risk they can't price. They lend, they take deposits, they need to know who's on either side of a payment. And the basic data fields in SEPA and Faster Payments are thin — first name, last name, and a transaction file. If a bank can't see anything else, it assumes the risk is uncapped. Because functionally, it is.

Crypto exchanges had the data, KYC, addresses, IPs, source of funds, but there was no clean way to share it through legacy rails. So banks defaulted to the safe answer: no. And the more they said no, the less they understood the industry, and the easier it became to say no the next time. A closed loop.

Shower thoughts: Product-market fit

It was just after June 2021. I'd just left my job as CPO at a big payments and BaaS firm. I was thinking about what to do next. And as cliché as it sounds, I was in the shower one morning when the pieces finally all snapped together. The personal pain at the bottom: every trader I knew had it. The institutional demand at the top: I'd watched it spike, week after week. These two things together represented something Y Combinator had drilled into me. What product-market fit actually looks like. It's not something abstract or counterintuitive. Rather, product market fit is when potential customers all but shout at you, begging for the product to exist. Clearly, both sides were shouting. The buyers existed, the sellers existed, yet nothing in the middle was working.

Another thing YC had drilled into me was: do only one thing and do it well. Most early-stage failures I'd watched were Swiss Army knives, businesses trying to serve everyone the moment they had a product. A startup has one shot. Spend it on one problem.

So that was the thesis. A B2B neobank for the crypto economy. Not crypto-adjacent. Not crypto-curious. Crypto, specifically, because the problem is specific, the regulation is specific, and trying to serve everyone would mean serving no one.

Signing without the money

A few months after founding the company, an opportunity came up to acquire Paybase, a UK e-money licensee that was winding down. They had the license, a couple of banking relationships, and a clean track record with the FCA. We negotiated a price.

There was only one snag: we didn't have the money. One of my co-founders at the time wanted to walk. And of course, it was a totally rational fear. Who would want to be personally liable for money they didn’t have, never mind for a brand new startup company that didn’t have funding?

But I signed the acquisition agreement anyway, with the hope that we’d be able to finance it from a funding round that hadn't closed yet. From the outside, that probably reads as reckless. From where I was sitting, it didn't feel like a bet. The demand was so loud, so specific, and so unmet that the only real risk was being slow. The round closed. The money went straight to the acquisition. Six months in, we had a license, banking relationships, and a beachhead.

And by March 2022, we had an MVP.

Our product: Gorilla Glass, not gold rush

From day one, we were positioned at the bottom of the stack. That’s intentional. The analogy I keep coming back to is Gorilla Glass.

Gorilla Glass sits on almost every phone you've owned. It's three or four steps removed from you in the supply chain: they sell to a factory, the factory sells to Apple or Samsung, Apple sells to a distributor, the distributor sells to you. You don't buy it directly. And yet you know what it is. You know it means quality. You'd notice if your next phone didn't have it.

That's the position we want. Player-agnostic, jurisdiction-agnostic, plumbing the fiat layer underneath whoever is building on top. We don't care if we're banking an exchange this today and a stablecoin issuer the tommorow. They all need fiat rails. We provide them, to a high standard. We’re neutral Switzerland, but in API form.

Four years in

The thesis hasn't changed. The sentiment around crypto has swung wildly. The cast of who's hot and who's collapsed has rotated. Stablecoins have moved from a curiosity to the centre of the conversation. Regulation was practically non-existent, and now it is crystallising all over. Yet none of that has touched the underlying problem we set out to solve: bridging TradFi and Crypto, at scale.

There’s a quote widely attributed to Churchill, despite there being no record of him ever saying it. Perhaps because it's exactly the kind of thing we can imagine him saying in his steadfast way: success consists of going from failure to failure without loss of enthusiasm. I think about that a lot. Four and a half years in, we're still pointing at the same problem we started with. The tank took a while to build. It started rolling in 2023. In 2025, when we hit EBITA profitability and default alive status, it started firing. Now, we reload and scale.