FoundationDB COO and Co-Founder, Nick Lavezzo, tells Startup Grind COO, Brian Park, how he continues to find the world’s best engineers in the DC area.

The longer I live, the more I relate to Danish philosopher, Soren Kierkegaard’s observation:  “Life can only be understood backwards; but it must be lived forwards.”

Looking back, Nick Lavezzo, an early employee of Visual Sciences and now Co-Founder and COO of FoundationDB, made one decision at 22 that changed the trajectory of his life.

Following His Instincts

Nick hadn’t set out to join a start-up.  A highly motivated student from the Virginia suburbs, Nick graduated with a BA in Accounting in 2002 from Oral Roberts University, the interdenominational Christian, liberal arts university in Tulsa, Oklahoma named after its evangelist founder.  He passed the CPA exam right after graduation and got a good job auditing clients like Walmart at the Tulsa office of a Big Four accounting firm.  One year later, Nick had the undeniable need to move back to the DC area where he had grown up.  He’d planned to complete a respectable few more years of experience at a local office of the firm.  When the Tulsa office blocked his transfer in an effort to retain the high-achieving employee, he quit and moved back anyway.

Riding the Wave with Visual Sciences

Just two months after Nick returned to Northern Virginia, his best childhood buddy from two doors down, Dave Scherer, and Dave’s friend from Virginia’s renowned “TJ” — Thomas Jefferson School for Science and Technology, Dave Rosenthal, invited him to join their early stage start-up, Visual Sciences.  There was nothing out there like this high-performance data visualization software platform and some big organizations had taken notice.

The timing was perfect.  The proliferation of data spinning out of government and fortune 500 companies meant that anyone with billions of records would want a product to create real-time, web analytics. “There were not many tools out there,” says Nick.  “Eventually, we owned the top tier of the market.”  “Doing a startup in 2004 was a lot harder because there were fewer tools to get your MVP [Minimum Viable Product] up and running,” he observes.  They tapped their strong, personal networks and built the company entirely with angel funding.

By 2006, Visual Sciences was sold to WebSideStory, a publicly traded web analytics company.  About a year later, WebSideStory was re-named to Visual Sciences after the Visual Sciences leadership team was appointed to manage the larger entity.  In 2008, it was acquired in a deal valued at $394 million by Omniture, and shortly after that the three friends left to take time off.  Two years later, the company was acquired by Adobe in a transaction valued at approximately $1.8 billion.  Nick’s proud of what he and his friends built: “Our platform is part of the Adobe Analytics suite now and I think it’s still rated by Gartner as the top web  analytics tool.”  They’d come a long way since the second grade when he and Dave Scherer’s first business venture was based on making soda more sugary.

Enter FoundationDB

“It’s great to have a win under your belt in the start-up world,” says Nick.  “But, personally, I knew I needed to keep doing cool new start-up stuff.  I was 28 when I was done with that company.  There was no question that as a team we wanted to continue doing valuable things.”  Now age 34, Nick and his co-founders are five years into building the company around FoundationDB,  a powerful, multi-model database that counts among its investors Silicon Valley’s Ron Conway and Sutter Hill Ventures.  It’s been a long road, but phase one, The Key-Value Store, is complete and downloadable as of last September.

Exploring One Idea Led to Another

After coming back together, the team explored new product ideas.  “We looked at Twitter and said ‘140 characters and somehow it’s worth a billion?'” Nick recalls.  The team decided to “churn out a bunch of easy applications, get them out the door and see what sticks.”

But, Nick’s two co-founders, being hardcore computer science engineers, wanted to make sure that they built the apps on a common stack so that the engineers could move between projects easily.  There were all sorts of great new technologies that didn’t exist in the early 2000’s and high productivity languages to use, but digging in to pick the data storage level of their stack presented the team with a compromise it didn’t like:

You could get a NoSQL database distributed across machines — a highly fault-tolerant design for the modern world of cloud computing — but these technologies gave up perfect data consistency or “ACID transactions,” something taken for granted for decades because of the popularity of single machine relational databases like MySQL or Oracle.

So, instead of building a bunch of easy-to-prototype apps, the team decided to build the powerful infrastructural technology they had wanted to use to power those apps.  They were on a mission to free up the world’s engineers by changing the way they develop applications for modern, distributed systems.

“We knew it was a really hard problem,” says Nick.  “A lot of people thought when we started this company that the trade-off  between perfect consistency and scalability was a fundamental, provable thing and that you couldn’t have both.”  When the folks they asked in Silicon Valley said they’d be interested in this product if it could be built, they added “sure, and a unicorn to ride around on would be nice as well.”

There Are Quicker Ways to Make Money

For almost two years, the co-founders weren’t totally sure that FoundationDB was going to be possible to build. “When we set off to solve this thing we knew we were going to have to take nontraditional approaches,” Nick says.  The co- founders spent the first year building development tools, including a new programming language, before they even started to build the database.  They bootstrapped the company with $1 million for the first year and a half.   “It’s definitely not a first startup idea,” said Nick. ” If we had wanted to build a company that would make some predictable money, we would have done a project that was lower risk and quicker.”

It was between this point and raising a $4.5 million angel round that the team started talking about their idea publicly, sparking epic Twitter battles, writing blog posts and articles.  They set out to convince people they were smart guys; thought leaders in the database world.  Until last week, their website’s biggest day was the first day their site trended on YCombinator’s Hacker News.

“There’s a ton of good angel seed and funds out there and you can’t have many degrees of separation between you,” Nick told the crowd at Startup Grind.  Hmmm… easy for him to say.  How did FoundationDB end up getting Silicon Valley’s legend Ron Conway to invest in their company?  The only thing more awesome than their prototype was their network:

 “We were not highly networked to the West,” explains Nick, “but we knew Howard Lehrman, CEO/Co-founder of YEXT, in New York. He’d gone to TJ high school around here with my two co-founders.  And, he was connected to the West and made an introduction to Michael Walrath, CEO and Founder of Right Media, an online ad exchange marketplace acquired by Yahoo for $850 million in 2007.  My co-founder, Dave Rosenthal went to MIT with a guy named Shawn who made the introduction to David Lee at SV Angel, Ron Conway’s fund.  That basically got us the lunch, but we also had the social proof.  They were investors in Yext.”

How did they find top-notch engineers in DC Metro area?  “We ended up being extremely surprised at how easy it was to find really good engineers here,” said Nick, who notes his team’s first hire was a MITRE engineer named Ben who they met on a walk while he was walking his kids around the old neighborhood where the team worked in the early days. Recruiter fees aren’t terrible in this town.   Of course, sending Inmail to engineers working at other government contractors via a premium Linkedin account works, too.

“There are people 10 miles from here working on problems of a scale that very few companies in the world will touch,” he observes.  “All that stuff that Snowden revealed — whatever you’ve decided on the morality — you’ve got to agree that it is impressive from a technical perspective.  Those people are probably around here.  They’re doing really cool stuff but it’s top secret and they can’t talk about it.  You won’t find these people in bars or the start-up ecosystem.”  To date, FoundationDB has hired 25 awesome engineers with zero turnover.  We’re still hiring and trying to grow 100% per year, and that’s a lot harder when you’re not three people.”

Part II of the Vision

“What we’ve built — I won’t say perfected, but gotten really, really good,” says Nick, “is a universal data storage engine that is distributed, highly fault tolerant and has perfect data consistency.  It has a relatively low level data model called an ordered key-value store.  A lot of our work right now is expanding the relevant audience of developers by adding new data models through these higher level data model translation layers so their favorite way of interacting with a database is available to them with FoundationDB.  The fact that each data model layer is backed by FoundationDB means that it inherits the performance we set out to build.”  Think building layers with Legos. That’s how they illustrate it on the FoundationDB website.

When asked whether or not they are concerned with competitors trying to replicate their progress, Nick replied: “This problem is not something that you can hire a bunch of people for and just tell them to do it.  Could you take a piece of Mozart and get a bunch of music grads and say write a new Mozart piece?  No, you need a new Mozart.”   “I’m sure people are working on it now, but we’ve been working on it for five years, and in five years’ time we’ll be even further along than we are now.”

One More Thing

One more thing that happened because Nick decided to move back to the area?  He met his lovely wife, Sarah.  The happy couple are just weeks away from going all in with the greatest start-up of them all:  Their first child.

Nick happy to advise attendees at a recent Startup Grind in Virginia.

Postscript:  Between Nick’s Startup Grind interview and this blog’s publication, FoundationDB announced Version 3 of its core product, which increases its scalability by 25x and makes it one of the fastest databases in the world.  This gained the company a top spot on Hacker News for 24 hours, and some great press including this Tech Crunch article:  http://techcrunch.com/2014/12/13/foundationdb-and-the-new-nosql/