Amkep Hub

CloudDeck AI Tech

Open models, reproducible experiments, fewer magic claims

ML engineer exploring small models, inference costs, and evaluation. Code-first explainers with benchmarks you can question.

Idris's hub

I build small, reproducible experiments to understand where AI models succeed and where they quietly fail. My notes focus on open tooling, deployment tradeoffs, and the gap between a convincing demo and a dependable system.

Here you'll find the models I use daily, the prompts that hold up under pressure, and short breakdowns of new releases worth trying.

ML engineer exploring small models, inference costs, and evaluation. Code-first explainers with benchmarks you can question.