My entire programming journey for the last 15 years has been me slowly climbing the static typing and functional ladder. I started my career as a C# dev. I was using TypeScript pre-1.0. Before I knew it, I was watching Strange Loop conference talks, finding random interviews with the creator of Elm, Evan Czaplicki , uncovering old talks from Rich Hickey, and keeping tabs on Richard Feldman in his work with the Roc Programming Language. At some point, it stopped being just professional growth, and how I defined "good" software, and a good programming language.
Which is why choosing Ruby on Rails in 2026 seemed like an odd choice.
I have written about how the #nobuild and HotWire approaches to simplified frontend architectures intrigued me;all while still getting hung up on the Ruby part of the stack. You know...a pretty significant detail. But finally I had an itch I wanted to scratch; a real app that I wanted to see exist in the world. As I explored different options, I thought this was the chance to finally give Ruby on Rails a fair shot.
Every language, every framework, every library, caries a mental model of how the world should work. Even if that model doesn't click with you, there is extreme value of trying it on. Seeing how it can change your perspective, or reinforce a previous one. I've watched DHH and his On Writing Software Well series. There is a way that he speaks about Ruby and poetic nature of the language and that comes from it. I wanted to steep myself in that mindset for a little bit. See if and how it would change my view of the world.
And surpisingly, I was shocked by how much fun I was having.
Rails has an answer for everything. That's it's whole thing. DHH calls this The One Person Framework. That idea never excited me much—until I wanted to end up with a real production app. I have enough experience to sling together my own Docker containers, write SQL, and stitch together a handful of services to cover everything I need. I’ve done that plenty of times. But it’s genuinely refreshing to have a first-party answer for every part of the stack without needing to assemble a bespoke solution.
Being so used to staticly typed languages, and a FP-mindset, I was taken back by the confidence I had with shipping new features. Obviously this didn't come from my normal comfort blanket of type safety, but strangely enough, convention.
For the past decade, the more complete a type system, the more confidence I had in what I was building. Without that safety, I assumed I’d feel the ground under me constantly shifting. Instead, I found that strong adherence to Rails conventions created its own kind of stability. It’s not the same safety you get from a Hindley–Milner type system, but it’s real, and surprisingly comforting.
At some point, I realized the fun I was having was building on itself, it was the reason the project didn't die like all my other side adventures. I bought a domain, I created a LLC, I created a business bank account, I completed security questionnaires. I had momentum. It became a real thing, with a purpose, and a name. PersonaliFI. A small, opinionated app for tracking net worth over time that didn't have unneeded weight or connection issues.
Challenging your own world view is so powerful. Although I still have an affinity for a good type system, I have found a new appreciation of a world I largely dismissed for the past decade and a half. What started as a fun way to challenge my long-standing views on programming languages, turned into my first product I shipped for real.