▸About
Who I am · why this blog · how to reach me.
Who I am
I'm an applied AI engineer working on LLM-driven systems — agents, RAG, and the production infrastructure that holds them up. Day to day that means reading papers, reading other people's code, and shipping systems that have to stay up. This blog is where that work gets written down.
Why this blog
I read and build a lot, and I'd rather externalize my notes than let them rot in a markdown drawer. Every essay here starts from something I actually needed to understand — a paper I had to evaluate, a tool I had to run, an architecture decision I had to defend.
A note on cadence: this blog started in May 2026 and ran at five posts a day for a while. That was a mistake — volume crowded out depth, and it read like what it was. Since October 2026 the schedule is three essays a week (Monday, Wednesday, Friday), and each one has to be something I'd still want to reference six months later.
What you'll get
- workshop · production lessons, code-first — code I actually ran, with the real output
- long-form · architecture & systems, with an argument rather than a summary
- paper · arXiv distilled with engineering bias — can I use this, and where does it break
- tools · honest reviews with hands-on comparison data
- weekly · the AI week in 5 bullets (archive, paused for now)
Original work
The code behind most workshop posts lives on my GitHub. If a post claims a number, it came from a run I did — not from a README.
Site history
- May 2026 — first post; static Astro site on Cloudflare Pages
- 2026 — 500+ essays across workshop, paper, long-form, tools, weekly
- October 2026 — switched from daily volume to three curated essays a week
EEAT note
Every article carries author and timestamps. Code I publish has been run; opinions I publish I've held for at least a week before writing them down. If I got something wrong, the contact page reaches me directly and I correct posts in place.