System design, architecture
and AI-assisted development.
Real decisions from a fullstack engineer who designs systems and ships them, increasingly with AI agents in the loop: why we chose it, what broke and what it cost. Nothing retold from the docs.
LLM Generation Parameters: Decoding Method Decides
Thirteen knobs sit in a watsonx.ai generation config, and one of them silently decides whether the three most famous ones do anything at all — and it isn't temperature.
Read the article →Tags
All tags →More articles
View all →
I Rebuilt One App's Architecture Three Times — Here's How I Choose Now
I rewrote the architecture of a single logistics app three times, each time chasing the "best" architecture instead of the one the project actually needed — and that mistake taught me more than any bo…
When `npm install` Becomes a Trap: The New Wave of NPM Malware
A logging library shouldn't be able to steal your AWS keys — yet that's exactly what the latest wave of npm malware is built to do, and it changed how I install.
Auth0 vs Keycloak: How I Choose for Scalable Identity
The hardest part of a distributed system isn't the microservices — it's deciding who owns identity, and that's the call I see teams underestimate the most.
Why I Keep Coming Back to Angular + NestJS + PostgreSQL
I didn't plan to fall for this stack — but a few months of shipping with Angular, NestJS, and PostgreSQL quietly turned me into a believer, and here's the honest why.