What it takes to ship an onchain credit card to non-crypto users

Devcon 8 · 3 Nov 2026 · 8:37

ether.fi engineer Yash Saraswat explores what it takes to deliver an onchain credit card to people unfamiliar with crypto. Drawing on production experience, he examines account creation and recovery, risk communication, payment failures, and the decisions that determine how much blockchain complexity users should see.

Speakers

Topics

From Devcon 8, hosted by Ethereum Foundation.

Transcript

Yeah. Hello everyone. So my name is Yash and uh I work as an engineer at Ether and uh yeah the title I use was a mouthful so I shorten it a bit. So today I want to talk about what it uh like how bringing non-crypto users onchain feels like and for that we first have to talk about like crypto UX. We know for many years that like crypto UX is has always been a problem. You need to manage your keys. You need to pay for gas. You need to keep track of multiple chains. And like even as a even as a developer myself, I sometimes feel this incredibly tricky to to deal with all of this stuff. And you can just imagine how difficult it could it would be for someone who does not know how to use crypto who has seen how

many hacks happen in crypto uh to and expect them to follow all these processes and actually like u use the crypto the way it is uh the way that we have been using for the past couple of years. An example of how how you would uh like we know like uh one of the biggest use cases of crypto is like crosswater payments for example but we know how hard uh it could be. Let's say the you have two people Alice and Bob and they want to transfer some money. First Alice has to ask Bob for his address. uh you need to copy that address, paste it into a wallet app and then Alice will if you if you know if you have good practices in crypto you will you will first send a test transaction confirm that the transaction actually goes through uh

because there are actually a bunch of hacks that can happen when you copy addresses and sometimes they like the hacker if they have access to your clipboard can change the address but yeah like no need to get into the details of that. Uh Bob can confirm his test transaction and then Alice finally sends the money. uh a lot of steps uh quite unintuitive and it's also using wallets managing keys as I mentioned like really really complicated we can definitely definitely do better than that the problem is not that crypto as a infrastructure is bad it's just that we haven't created the the UX required to bring uh non-crypto users on chain so what we can do we can build a proper unified app we can build a define

neo bankank where everything is consolidated ated where the user does not have to think about all these problems where you don't have to think about where the money comes from in the first place. There is native banking and on-ramp offramp integrations. You can spend using uh your crypto. You can you can connect with the Visa network, create credit cards for users and allow them to tap and pay with their crypto money. You can uh integrate all crypto-related features in the app in a way which feels like you're using just the regular banking app. You can uh send people money using tags. You can send people money using payment links. You can borrow against your assets. You can uh buy stocks. You can buy gold. You can buy crypto. You can do everything in a single app. And you don't need to manage

gas ch you don't need to manage like chains. You need to you don't need to manage your keys and all that stuff. And all of this makes the user experience uh far better. And uh the next question you might ask is that oh so like we're just like rebuilding banks from scratch but just using crypto underneath. What's the benefit of that? The major benefit is that uh with a with a defa bank you get uh non-custody. In a non-custodial architecture, each user's funds stays inside their own smart contract account. Uh you load crypto, it straight up goes into your u uh uh into your wallet and you and you don't control the keys. You can uh you

can use a service like preview or you can use a service like turnkey like we do uh to manage the keys for the users. But uh that does not give uh and uh and and solutions like turnkey for example use secure enclaves so that even turnkey and even we at etheri do not have access to your keys. Only the user can create a secure session and uh and access their funds. So they don't have to worry about the keys. They don't have to worry about us stealing their money. They don't have to worry about uh turnkey stealing their money. If etheri for example goes bankrupt tomorrow then uh in our bankruptcy filing like the users would not be uh would not be subject to like whatever the filing is because like it's uh uh it's non-custody users are actually the ones who are holding their

funds and we don't control it at all. So the user funds are safe uh uh with respect to like whatever happens to ether or turn key or anything like that as long as uh this the architecture is secure. Obviously as for card spendings like u yeah as I said like the uh the user can only initiate a secure session with turnkey by using uh by using pass keys or by using OTPs which means that we don't have access to the user funds and we can only spend user funds when they actually do a transaction but even there users can uh set onchain limits to the amount of money that we can u spend on their behalf and even when we spend on the behalf like uh the way our smart context is set the money only goes from like

their safe wallet to an account that we have from which it can either go to a rain dispatcher address or it can go back to the user. So there is very limited ability of what we can do and one of the best things about having like a bank with the crypto with crypto rails is that you have access to so many financial instruments that you don't normally have. For example, as I mentioned there is easier easier crossber payments. I know like uh any anyone who has tried remittances with a traditional bank knows how shitty it is.

The best thing you can do is you can borrow against your assets. You can hold gold, stocks, um, and any crypto asset and like spend against it. You can trade outside market hours. Uh, and you can also use permissionless payment links to um uh to uh, to like send money to your friends. And, uh, all of this uh, like there's proof that actually using good crypto UX actually brings users on chain. We have, for example, Ether has done more than $1 billion in card spends. We have more than 35,000 weekly active users. And we have done more than $300 million in bank transfers. Uh which is like proof that if you actually improve the UX uh UX of using crypto, people will actually start

using it because of all the benefits that I I had mentioned. And uh yeah, that was basically a brief about uh >> thank you so much Yash. And we have a time for maybe one question. >> I will here I'll give it to you. How many of those 35,000 users are non-crypto users? uh like I would not know how many of them are non-crypto exactly but like they're basically using this product as their day-to-day as a day-to-day card like we can see the statistics they're using it to buy like coffees they're using it to buy everything >> they have crypto assets because it needs to be backed by crypto right >> uh no because we have on-ramp and

off-ramp solutions so as I said like we have I think out of the 300 million bank transfers I think around 100 millionish are like around on-ramps so you can assume assume like uh most of those uh people who have on-ramp like fiat money into their ether account are uh like probably not crypto users because like they did not uh add crypto directly they used a banking rail. So I can assume that uh a bunch of them are definitely non crypto users. All right. >> Okay. So, uh two questions, right? One, uh how do you take care of uh uh AML KYC uh at the back end from a regulatory standpoint? And second, which are the

counterparties that you use uh for all of these fiat conver uh conversions. >> Uh uh yeah. So to answer the first question uh we basically use uh uh we partner with rain for the entire transaction uh infrastructure. So they basically manage the KYC around the uh so we basically do the KYC using sums I think uh we collect the the user KYC and then we like forward it to rain and rain is the one who manages the KYC for card issuances and then to answer your second question we have a bunch of integrations for banking rails we have due bridge uh and we recently partnered with moon pay as well so we have a bunch of options and these providers basically allow us to serve as many users as possible with

like multiple currencies multiple options onboarding options and yeah etc. >> Thank you.