Naseem.
Hi, I'm Hamza, based in Karachi, building AI systems for businesses globally.
I focus on one problem: repetitive work that humans shouldn't have to do. Right now that's mostly voice AI that answers clinic calls, books appointments, and follows up with patients, around the clock.
Before I wrote a line of code, I spent years studying classical Islamic logic, Ibn Sina's syllogistic frameworks, specifically. That's not a detour. That's why I think about AI reliability differently than most developers. I don't settle for systems that work most of the time.
I co-founded Bytelytic. Published research on LLM verification, LogicGuard, 100% precision, zero false positives. Selected as a Section Leader at Stanford Code in Place 2026.
If you're building something in AI, or you need a system that actually works, not just demos well, I'd like to talk.
Stanford, published research, and the work that earned the credibility.
What I've been shipping lately - pulled from GitHub.
Things I built end-to-end.
Building AI systems, leading teams, and freelancing - the work that shaped what I know.
What I ship with. Built up over a few years - mostly stable. I switch slowly and only when something is clearly better.
Notes from building. Mostly AI, logic, and engineering-flavored.
Reading classical Islamic philosophy - Ibn Rushd, Al-Farabi, Ibn Sina. Running. Thinking about why AI systems fail where human reasoning doesn't. Building things nobody asked me to build yet.