Inside BlockCypher's plans to become the "AWS of blockchains"

Profile picture for user jreed By Jon Reed July 21, 2016
One of the memorable blockchain startups I interviewed this year was BlockCypher. Co-founder and CEO Catheryne Nicholson doesn't lack for ambition. But how did her team find their way to the enterprise blockchain, and to filling a gap in blockchain cloud infrastructure? Here's their story.

If blockchain becomes a force in the enterprise, the heavy lifting will be done by those who can fuse blockchain tech with enterprise scale and sensibilities. That's where BlockCypher enters the picture. During our sit-down at Coindesk's Consensus 2016 event, BlockCypher CEO and co-founder Catheryne Nicholson told me how an encounter with Bitcoin technology convinced her that a blockchain-for-enterprise play made sense.

And with her team's cloud architecture chops, building what Nicholson calls "The only cloud-optimized blockchain technology layer" was the logical move. I know what you're thinking: aspiring to be the "AWS of blockchains" is aspirational enough to flirt with delusion, but I'm partial to startups that think big. And it's not like BlockCypher is a hypothetical idea; their platform is live and in use. They currently process five million requests a day.

The origins of BlockCypher, and Bitcoin's performance problem

BlockCypher's home page is plastered with customer logos, and while many of them are crypto-currency related, their platform is blockchain-agnostic. Deloitte is one of BlockCypher's key partners as they pursue their enterprise ambitions. I asked Nicholson how it all began. The roots of BlockCypher go back to 2009, when she and BlockCypher co-founder and CTO Matthieu Riou were working on a prior mobile app collaboration:

We were building a mobile web app, but we kept running into: "What is a substitute for cash and check?" There is nothing. The Internet was designed to share information on a massive scale; it was not designed to transfer value.

That led them into a Bitcoin evaluation:

My co-founder had mined Bitcoin. He sent me some; I tried out all the wallets. They were just terrible, it was nothing I could retire on, but tinkering with it was interesting.

The "double payment" problem Bitcoin solves made an impression:

Bitcoin achieved something that had never been done before, which is consensus on a massive scale, and also the ability to transact value. The reason I cannot take a picture of a dollar or check and send it to you via an attachment in an email is because you cannot prove that I haven't sent it to a million other people... Blockchain technology fundamentally solves the double payment problem.

Nicholson tried to make an enterprise connection:

It was very clear to us that this type of technology was extraordinarily disruptive, but where it was going to go, we had no idea.

Nicholson at Consensus 2016

Maybe they could find a niche that played to their strengths:

We have over thirty years of experience bringing and building enterprise applications to market, but we were trying to think, "Where do we fit in to this whole thing?" We are not a payment processor, we are not finance deep, Our core competency is in core enterprise infrastructure.

A Bitcoin test on a laptop confirmed the scale problem:

At the end of 2013, we took a look at what was then Bitcoind, which is now Bitcoin Core, a reference client that underscores Bitcoin. I tried running it on my laptop; it ground it to a complete and utter standstill.

For Nicholson, seeing this clunky software was an "aha" moment:

What it has been able to achieve so far is transformational, and yet there is no way that the enterprise can come in - or even a large-scale startup - and build on top of something like this.

Supporting enterprise-ready blockchains

So, in 2014, BlockCypher was launched. They redid the Bitcoin architecture, but made it blockchain-agnostic:

We rebuilt it from the ground up, and implemented the entire Bitcoin protocol, but made our infrastructure agnostic to any Blockchain. We saw the potential of where it was going to go. We didn't think this would just be one chain; it was going to be a multitude of chains. That is how we designed infrastructure.

Because Ethereum was big at Consensus 2016 for running enterprise blockchains, I asked Nicholson about that, as well as Hyper Ledger, which several readers of mine are tracking:

We support Etherium, yes. If Hyper Ledger ever comes to production - whatever blockchain needs to run at scale. There are hundreds and hundreds of blockchains out there. Which ones we support are the ones that get traction. [Editor's note: see BlockCypher's July 22, 2016 announcement of their support for the Ethereum contract API).

Having been through the pain of on-premises-to-cloud, BlockCypher built their platform on cloud:

We knew that that cloud infrastructure piece of that is going to come and play with blockchain technology. When we built our infrastructure, we made it fully cloud optimized.

BlockCypher's platform has been up and running for more than two years. When I talked to Nicholson, they were celebrating the one year anniversary of their first live customer. So far, they've been reliable, with an uptime of 99.999 percent.

So how did they prove out their use case? By taking their solution directly to developers:

We would talk to people that were building applications to say, hey are you running Bitcoind? Would you like to try out our APIs? The reason we call April 25th our anniversary is because Adam Draper, who would later on go on to invest in us, was running a Boost PC Bitcoin hackathon that weekend. At the hackathon, the winner and the third place team used our APIs.

And voila, their first proof point:

Adam asked them, "How are you able to build this in a period of 24, 36 hours?" It was because they had utilized our APIs. That was the first proof point, that instead of spending three to six months building out infrastructure before you can then just do basic things, that if you use an API like BlockCypher, you can get out the gate in just a couple of hours.

About that AWS thing - what tools do blockchain developers need?

In the same manner that AWS has a wide array of services for developers, we have an array of web services, APIs, web hooks, web sockets, which a developer can use.

I asked the expected question: security. BlockCypher provides a multi-signature API, which allows co-signers. Double-spend notifications protect against theft and loss. "It's very difficult to build multi-signature security from scratch," asserts Nicholson. Not being a developer I can't personally vouch for that, but it does sound like a pain in the tuckus. "Now you can just use that API to enable multi-sig in your wallet, or in your app. That's just one of our services," Nicholson says.

Is the enterprise blockchain gaining momentum?

One twist on the whole AWS-for-blockchains tagline: currently, BlockCypher primarily runs on AWS themselves. I asked Nicholson what gives BlockCypher the edge over other blockchain service providers? She didn't hold back:

We are the only ones that can post those reliability numbers. No one has the breadth and depth of our services. We do everything from an open asset API, to a transaction wallet API.

Best-in-class aspirations are a good thing, but BlockCypher's success does depend on enterprise blockchain adoption. So - where are we? Nicolson pointed to announcements from BlockCypher's partner, Deloitte, as one example of the enterprise momentum - including five partnerships and twenty blockchain prototypes (update: there has been plenty of Deloitte blockchain news since we talked also). Nicholson:

We have executed a teaming and partner agreement with Deloitte because they bring in the industry expertise, the consulting services. We bring the raw technology. Deloitte doesn't go after startups; they go after enterprises.

Nicholson pointed to the presence of IBM and Microsoft at Consensus 2016 as another sign of enterprise momentum. She wasn't concerned that the financial services firms weren't present at Consensus 2016 - "They tend to have their own conferences."

The wrap - identity and other use cases

We wrapped with a look at use cases. Are the use cases mostly financial services?

Yes and no. We see the whole variety of use cases from the start. In a way, it is the canary in the coal mine, because the start ups are the ones that tinker and experiment. Our customers include ATM providers and gaming companies. The identity folks are our customers. These are all the same things enterprises will be looking at - on a much larger deployment.

Identity was cited frequently during my Consensus 2016 interviews. Nicholson has seen the interest firsthand:

Identity is huge. ShoCard is a really great customer of ours. The founders used to lead identity management at Yahoo. Their core competency is digital identity. They came to us, and said, "We will use you guys as the infrastructure provider, and we are focused on how we manage that mobile, digital identity. Every single enterprise, every single bank, every single insurance or energy company or whatnot - they all have customer databases. They are all trying to figure out: are you truly a person, instead of some made-up person that is trying to hack the system. That is applicable to almost every single enterprise.

It's still early days in BlockCypher's plans for blockchain domination, so we'll have to stay tuned. They completed a large seed funding round in January 2015, which included tech investor luminaries. Now, I said, it's time to kick ass and execute. Nicholson nodded in agreement.