Express Cloud crossed 2 billion naira in business assets recorded and managed through the system. I was happy about it, of course. But the funny thing is, the number made me think about medical school.
I spent two years studying medicine before leaving for computer science. It felt like a big risk then. Medicine had a clear path; software was something I loved, but I couldn't have promised anyone it would work out. I worried about starting over and whether I was throwing away two years.
Looking back now, I'm glad I left. Not because medicine isn't a good career. It just wasn't mine.
Recorded business assets managed through the system. Not money I own, and not the software's valuation. Real operations that need accurate records.
Small fixes, big lessons
September was a lot of reading code I didn't write. I contributed two correctness fixes to an open-source trading system, and both were merged. One was a boundary-condition bug; the other involved state that didn't properly reset when conditions changed. The fixes were small, but getting them accepted meant understanding the existing code, writing the right tests and listening to reviewers.
That feeling stayed with me. In October, two frontend fixes for an open-source sign-language AI project were merged too. Other contributions are still under review, which is fine. I like the fact that open source doesn't care how excited I am about a change. The work still has to hold up.
And then there was the actual job
October was busy. I worked across financial workflows, mobile experiences, booking flows, uploads, security reviews and a lot of UI details that looked tiny until someone tried using them. I also learned to stop pushing half-finished batches of work just because a commit was ready. CI time belongs to the whole team.
Express Cloud was growing too. More sales, more operational demands, more reasons to be careful. We upgraded the server in preparation for expansion, checked the data, and made sure the background work and backups were running. The satisfying part wasn't the new server. It was getting to the point where people could keep working without having to think about the move.
We observed two sales in the same second. The retained web logs also showed three requests in flight simultaneously. That's not a historical concurrent-user figure or a benchmark to boast about. Just evidence of a system being used.
I bypassed my own release pipeline.
Yes, the one I built to stop risky releases.
There was an urgent fix. I took a shortcut around the usual checks, and the application broke. Omo. We recovered, but I knew exactly what I'd done wrong.
The lesson wasn't complicated. If the release process is too slow, improve it. Don't bypass it and hope the shortcut works. I could give this incident a dramatic postmortem title, but really, I made a bad call and learned from it.
Still learning, still figuring out rest
I'm also studying for an MSc in Financial Engineering at WorldQuant University. Recently that has meant liquidity, regulation, securitization and risk. Some evenings I've gone from troubleshooting a real system straight into a lecture about financial systems. It can be exhausting, but I enjoy seeing the concepts connect.
I spent time writing too, mostly about what happens behind the screen: AI, infrastructure, commerce, identity and payments. It turns out a lot of my work comes back to the same question: is the system telling the truth when somebody depends on it?
And yes, I enjoy being paid well to do this. It's one of my favourite things about choosing engineering: I get to earn a good living doing work I genuinely like. But I am also learning that liking my work doesn't mean I should be doing it every waking hour. I'm trying to sleep more, be present elsewhere and stop treating every free evening as an invitation to open another project.
Connecting the dots
Steve Jobs said this in his Stanford commencement address:
I couldn't see the connection when I left medicine. I certainly couldn't see these two months: software handling real business operations, code accepted into projects I don't own, graduate study and the occasional very public reminder that I still have things to learn.
I can see a little more of it now. September and October were not perfect, and I'm not pretending to have work–life balance figured out. But I enjoy what I do, I am getting paid to do it, and I'm glad I took that risk.
Omo, we move. Preferably after some sleep.
Notes and references
