Technical deep-dives, debugging stories, and infrastructure chronicles
How strategic context restriction dramatically improves AI agent effectiveness. Discover why giving AI agents MORE context often makes them perform WORSE, and how to fix it.
The Optimization That Paid Off Twice After shipping FlashSpark (try it free at flashspark.eddykawira.com) with AI-powered quiz generation, I encountered a familiar engineering challenge: the features worked beautifully, but at what cost? Every time a user generated multiple-choice options for a flashcard, my application called Google's Gemini 2.5 Flash Lite API. At $0.10 per million … Continued
I had a Claude Code skill that worked perfectly — and silently burned ~7,300 tokens every run. The issue wasn't logic. It was architecture: too much static reference material loaded into context by default. By switching to progressive disclosure and moving heavy reference logic out of always-loaded markdown, I kept the same outcomes with ~500 … Continued
Running production-grade services from a homelab sounds risky—open ports, DDoS exposure, residential IP reputation issues. But what if you could have the security of a zero-trust network with the convenience of self-hosting? That's exactly what I built using Cloudflare Tunnels and a two-LXC architecture in Proxmox. This post walks through the architecture powering this very … Continued
How I enhanced an open-source DNS sync tool with real-time monitoring and deployed to production in a single day. Go backend, JavaScript frontend, zero new dependencies.