Skip to main content

Multiple Developers and a Credit Platform

Updated 4 min readMarkdown

Multiple developers are enough to build a real platform. They are also enough to build multiple different platforms if nobody holds the center.

At MonitorBase, where I was Chief Technology Officer (CTO) and later VP of Development, we ran credit monitoring for 10,000 clients, with more than 7 million queries a month. My team was multiple developers. The promise to clients was simple to say and hard to keep. A client logs in and the dashboard feels instant.

Instant is not a mood. It is work.

What the client actually felt

Clients did not care about our stack. They cared about three things. Is the data there. Is it right. Does it load now.

That set every priority. A feature that made a dashboard slower had to earn the slowdown, in public, with numbers. Most did not. A report that was accurate but late was still a problem, because the client made decisions off that screen.

We built in Laravel. I have more than 10 years in Laravel in production. A team moves faster in a framework it knows cold than in one it is still learning while clients wait. Boring familiarity is a leadership choice.

We used Redis heavily around the product so the logged in experience stayed fast. Credit data itself was never cached. That line was compliance, and every architecture call after it had to respect it. One clear line like that saves a hundred arguments later.

Leading is removing arguments

My job was not to be the best developer on the team. On a team of that size, that claim stops being useful fast, and it is usually not true on any given Tuesday.

My job was to remove the arguments that waste a week.

What are we building. What are we not building. Who owns the call when the tests disagree. What number decides it. If those four questions have answers, good developers do good work without a lot of supervision theater. If they do not have answers, you get meetings about meetings and code that reflects the confusion.

I wrote before about the SingleStore migration and what it did to our slowest jobs. The leadership part of that story was not picking a database. It was letting the team run the evaluation on our real workload, then making the call and not reopening it every month. A decision a team helped test is a decision they will defend at 2 a.m. when something breaks. A decision handed down from a vendor slide deck gets relitigated forever.

That pattern held on smaller calls too. Set the criteria in the open. Test the real thing, not the demo version. Make the call. Write the number down. Move on.

The part talent does not fix

Talent does not fix unclear ownership. I learned that the expensive way, more than once.

When two strong developers both think they own the same piece, you do not get twice the work. You get two versions and a merge problem. When nobody thinks they own a piece, it rots politely until a client finds it.

So I got blunt about ownership. One name per piece. That person could ask for help and change their mind with evidence. What they could not do was leave the rest of us guessing who was driving. Clear ownership is not micromanagement. It is the opposite. It is how you leave people alone and still know the work is covered.

Protecting focus worked the same way. A platform serving 10,000 clients generates a steady stream of urgent sounding requests. If the team chases all of them, the important work never finishes. Part of leading was saying no, or not yet, and taking the heat myself instead of passing it down the line.

What I use now

The scale is different now. I run Eco Moving Utah, a moving tote rental business in Salt Lake County, and I build Bin X, the bin rental software that runs it. The team is mostly me, with an AI agent doing the first pass on a lot of the work around my calls.

The lessons carried over clean. Eco Moving Utah customers do not care about my software either. They care if the totes show up on the right day. Bin X

About the author

  • Guy Warner is a Salt Lake City software engineer with 20+ years building PHP/MySQL/AWS SaaS products, with 10+ years of Laravel in production. He was Chief Technology Officer (CTO) and VP of Development at Monitorbase in credit monitoring, handling 7M+ monthly queries, where he led 12+ developers and migrated to SingleStore. He founded Eco Moving Utah, a moving tote rental business, and builds Bin X, rental business software. He has a B.S. in Political Science, University of Utah.

No comments yet. Be the first to comment.

More from this blog