Writings

2026

The Claude Code Tabs You Forget About I kept losing Claude Code sessions that were waiting on my reply. So I built a tiny local dashboard that surfaces them. May 23 · 3 min read Meta-Dev-CLI: Stop Clicking, Start Typing Why I built a CLI for Meta's WhatsApp API to keep developers (and AI agents) in their flow state. April 2 · 2 min read Live Kitchen Feeds: Transparency Without the Cloud Bill Why food delivery apps need live streams, and how to build them cheaply with WebRTC. March 31 · 3 min read Designing for Scale Scale is mostly about clarity — of data flow, ownership, failure handling, and observability. March 12 · 6 min read Event-Driven Systems Are Not Magic Event-driven systems are powerful, but they redistribute complexity rather than eliminating it. February 20 · 7 min read Debugging Distributed Systems Debugging distributed systems is fundamentally different. You're reading timelines, not stack traces. February 5 · 6 min read Production-Ready AI Adding AI to a system is easy. Making it reliable is not. January 18 · 5 min read Why Queues Matter More Than You Think Queues are the unsung heroes of reliable systems. They absorb traffic spikes, decouple services, and turn fragile sync calls into resilient pipelines. January 5 · 5 min read

2025

Boring Systems Win The most reliable systems are boring. They use proven databases, avoid unnecessary abstraction, and log properly. December 10 · 4 min read Data Flow Is Everything Every backend problem eventually becomes a data flow problem. November 22 · 6 min read

2024

Elasticsearch Sharding, Done Right Sharding and index mappings are the foundation of a healthy Elasticsearch cluster. Get them wrong, and no amount of hardware will save you. November 10 · 7 min read

2023

You Probably Don't Need Microservices The microservices hype led teams to split too early. Most systems would be better off as a well-structured monolith. June 15 · 5 min read