I just shipped v1.1.0 of oh-my-kimi — a multi-agent orchestration harness that wraps the Kimi Code CLI (K2.6) into parallel coding teams. One prompt → planned, parallelized, reviewed project: npm install -g @oh-my-kimi/cli omk chat — Interactive Kimi session with resumable context, tmux support omk cockpit — Real-time dashboard with parallel TODO/agent rendering omk hud — Full terminal dashboard
Every developer has been here. Debugging why Puppeteer crashes in Docker but works on your machine And you still haven't built the actual feature you needed the PDF for. So I built Templar Describe your document Tell the AI what you want an invoice, a report, a contract, a receipt. It generates the HTML template for you. Call the API with your data POST /api/render { "templateName": "invoi
I wanted to add live chat to my WordPress sites without loading a 500KB third-party script. So I built my own. GhostChat is an open source embeddable Widget: Vanilla JS, no framework, ~10KB Backend: Cloudflare Workers + Durable Objects for persistent WebSocket connections Payments: Stripe for the hosted tier Self-hostable: Bring your own Cloudflare account Durable Objects give you stateful serve
The more I use AI, the more convincing it feels. Clear answers. Whether it’s: strategy code writing decision support AI rarely hesitates. And over time, I noticed something subtle. I stopped questioning it as much. Breaking the Expectation We assume better tools reduce errors. Smarter systems. And in many cases, that’s true. But there’s a hidden shift happening: As AI improves, our skepticism decr
Most candidates overthink "Tell me about a time you failed." They assume the safest move is to soften the story, pick a harmless mistake, or package a "failure" that is secretly a strength. That usually backfires. In software interviews, especially for experienced engineers, a real failure is often better than a polished non-answer. Hiring managers are trying to figure out whether you can own mist
Blueprint Felonies Software isn't a puzzle to solve; it is a liability to be managed. In high-stakes, cloud-native environments, the line between "sophisticated" and "unstable" is razor-thin. With over 17 years in the software trenches, I’ve seen architectural "thinking mistakes" destroy more careers than bad syntax ever could. We often build massive, intricate systems when a simple, focused sol
Find a beginner-friendly issue. Fork the repo. Set up the dev environment. Read through the codebase. Start working. Then check the issue again and see a comment from 2 days ago: "Hey I'm working on this, should have a PR up soon." Two hours wasted. Every single time. The weird part? Almost every existing tool for finding open source issues - goodfirstissue.dev, up-for-grabs.net, codetriage - rely
Discord rewrote their stack multiple times. Elixir for real-time messaging. Python for APIs. Go for microservices. MongoDB for storage. Electron for desktop. Standard startup choices. Ship fast, figure out the rest later. 5 million users. MongoDB couldn't keep up. Switched to Cassandra. 12 nodes. Worked fine. Until 2022, when those 12 nodes became 177. Maintenance got painful. Costs climbed. They