About

I ship AI systems for a living. I write about the gap between the demo and the deploy.

What I work on

I'm a Data Scientist and AI Engineer at Bio-Techne, where I lead GenAI initiatives across the company. I've shipped a multi-agent sales intelligence platform serving hundreds of weekly users, a production technical-support assistant retrieving over a million Salesforce records, document intelligence pipelines that process PDFs at scale, and decision agents that have to answer correctly the first time because there isn't a second chance.

Before GenAI was a category, I built classical machine learning systems — recommendation engines, predictive pipelines, anomaly detection — and ML pipelines for protein engineering that commercialized AI-designed therapeutics. The tools change every six months. The hard parts don't.

What I write about

Most public writing about AI right now is either hype or theory. There's a shortage of writing from the people actually shipping these systems into regulated, high-stakes environments where "the model hallucinated" is an incident, not a tweet.

This site is my contribution to closing that gap. Specifics over generalities. Scars over claims. The things I wish someone had written before I had to figure them out the hard way.

Topics I keep returning to: grounded retrieval, evaluation that survives reality, agent architecture, production failure modes, and the long unglamorous middle between a demo and a system you can run at three in the morning when something breaks.

Get in touch

Email: hello@ilyashassan.dev
LinkedIn: /in/ilyasalhassan
GitHub: @ilyas-hassan