Not a typical origin story.
person
behind it ←
I came to computers late, dropped out of college, and spent a good while not being particularly good at anything. A friend suggested I try software. That was the whole turning point.
Before that, I was drifting. After that, I had something to show up for. I don't think I'm naturally gifted — I just found the work that makes me feel like I'm in the right place, and I've been building ever since.
Five-plus years in, I've shipped insurance platforms, pharmacy systems, logistics dashboards, component libraries, and a few things that don't fit a neat category. These days it's mostly Next.js and React 19 — sole developer on ChiNet's site and CMS, 384 of ~550 commits on their logistics platform, and a 236K-line legacy system rebuilt on Vue 3 in two weeks. A lot of it has been high-stakes Ethiopian enterprise — but that's context, not a constraint. I'll build whatever's worth building.
- Stack
- Next.js 16 React 19 TypeScript Node.js Prisma TanStack Tailwind Vue 3 PostgreSQL MongoDB Docker etc.
- Also in the loop
- Claude Code, daily CLAUDE.md project rules MCP servers Vitest + Playwright
- I care about
- Systems that hold up Teaching what I know Finishing things
- I'm tired of
- AI fear mongering A new JS package every week (I'm part of the problem 😅) — not anymore though, since AI came to the picture it seems like open source is about to die
Full-stack apps — Next.js end to end
Architecture to production: App Router and Server Components on the front, Prisma data models, auth and Docker deployments behind them. Component-driven from the start, so the codebase doesn't become a haunted house six months later.
Component libraries & design systems
Custom, reusable component systems — tables, forms, modals — built for your exact use case, not for the npm average. Published one. Built several more that didn't ship to npm but should have.
Healthcare & fintech UI
Complex domain, high stakes, zero room for confusing UX. I've built for hospitals and insurance companies — the kind of software where a confusing modal is actually a problem.
Technical leadership
Code reviews, architecture decisions, onboarding new devs. I've led frontend teams and lived to tell the tale. I'll speak up when something is heading the wrong direction — before it becomes a problem, not after.
Listen first.
30-minute call. I ask too many questions. You walk away with a written summary of what we agreed on, either way.
Something clickable.
No wireframes, no Figma decks. A real prototype — you interact with it, you tell me what's wrong, we fix it.
Ship in slices.
Weekly check-ins. Always deployable. You always know what's done and what's next — no black-hole sprints.
Hand it over.
Docs, walkthrough, two weeks async support. Then I get out of your way — the goal is that you never need me to explain your own codebase.
One of the best engineers I've worked with. Great instincts, clean code, ships on time. Genuinely impressive human being. I'd work with him again immediately. I am completely unbiased and this is a real quote.
I wrote all of these. Every single one. This testimonial section is fabricated. But the work in the projects section is real, and the code on GitHub is real — so maybe just check that instead.