Why a blog?
I have been meaning to start writing for a while. Not to publish polished essays, but to have a place where I can think out loud — about things I am building, approaches I am experimenting with, and ideas that do not fit neatly into a README or a LinkedIn post.
The name of this site already says it: progress, not perfection. So here goes.
What to expect
Most posts will be technical. Topics I expect to come back to again and again:
- Edge AI and embedded systems — running inference on constrained hardware, real-world tradeoffs
- Image and signal processing — the kind of problems that are harder than they look
- MLOps and tooling — getting models from prototype to production
- Home lab and self-hosting — because tinkering at home is how I stay sharp
Occasionally something less structured — a tool I discovered, a mistake worth documenting, or a rabbit hole that turned out to be worth it.
A note on format
Posts here will be short. I would rather ship something readable in an hour than spend a week on something perfect. If a post is useful to one person besides me, that is a win.
Setup
This blog is built into my portfolio site. Posts are plain Markdown files — no CMS, no database. Just files, marked, and a Vite prerender pipeline that generates static HTML at build time. The source is on GitHub.
# New post = new file
touch src/posts/YYYY-MM-DD-title.md
That is the entire authoring workflow. Simple enough that I might actually use it.