Same Skill, Six Agents, Nine Models: What a Real Eval Matrix Taught Me
I ran the same skill and the same task through different coding agents. The model mattered. But the agent harness changed the outcome far more than I expected.

Search for a command to run...
I ran the same skill and the same task through different coding agents. The model mattered. But the agent harness changed the outcome far more than I expected.

A major international takedown disrupted hundreds of systems. The StealC infrastructure I documented still answers on its known malware routes.

How a simple rationale-capture skill evolved into a repository-native knowledge lifecycle without a database, daemon, or dashboard.

I turned four months of working with AI coding agents into an open-source skill for preserving the reasoning behind a codebase.

Hashnode serves the correct sitemap when you add one small query parameter.

An open, just-for-fun prediction challenge for everyone using ChatGPT, Claude, Gemini, Grok, local models, or any other AI.

Using UBDCC on six low-cost Vultr nodes to synchronize 4026 replicated Binance Spot and Futures DepthCaches, monitor the cluster, and push the REST API with Grafana Cloud k6.

A step-by-step guide to running your own Binance order book service on Kubernetes — live, always-in-sync market data served to any app, in any programming language, over a simple REST API. Works on any Kubernetes provider, including Vultr, OVHcloud, Google Cloud and others.

REST, WebSocket, order books, trailing stops, and cluster-scale infrastructure — the production-grade Python stack for Binance.
