15 Years of Building for the Web

I wrote my first line of production code in 2009. It was PHP, it was terrible, and it powered a feature that thousands of college students used daily. That tension between messy code and real impact has defined my career ever since.

The startup years

CampusLIVE started in a UMass dorm room. We built a platform for brands to connect with college students, raised $3.1M from Highland Capital Partners and CRV, and scaled to dozens of campuses. I was employee number three.

What I learned: startups don’t fail because of bad code. They fail because they build the wrong thing. The best engineers I worked with were the ones who pushed back on requirements, not the ones who implemented them fastest.

After CampusLIVE, I joined CustomMade, a Google Ventures-backed marketplace connecting artisans with customers. Different stage, different problems. At CampusLIVE, the challenge was building fast enough. At CustomMade, it was building things that scaled: search systems, matching algorithms, payment flows that handled thousands of transactions.

Going independent

In 2015 I went freelance. Five years of working with Boston-area startups across fintech, healthcare, and consumer products. Every engagement taught me something different about how teams ship software.

The best clients gave me a problem, not a spec. The worst gave me pixel-perfect mockups and got frustrated when I suggested a simpler approach that would ship in half the time.

Going independent also taught me that technical skill is table stakes. The engineers who thrive are the ones who communicate clearly, manage expectations, and know when to say “this is a bad idea” diplomatically.

Building VisualBoston

In 2020, I co-founded VisualBoston with a designer I’d worked with for years. The premise was simple: senior talent, no overhead, available when you need us.

We’re not trying to be a big agency. We’re two people who’ve shipped dozens of products and know how to move fast without cutting corners. Our clients get the same people from kickoff to launch. No handoffs to junior developers, no “account managers” who don’t understand the code.

The agency model lets me work on diverse problems while still building my own products on the side. TallyQuote (an all-in-one platform for contractors) and GardenMates both started as side projects that grew into real products because the agency gives me the financial stability to take risks.

What I’d tell myself in 2009

Ship it. The code doesn’t have to be perfect. The architecture doesn’t have to be elegant. The feature has to work, and the user has to care. Everything else is negotiable.