MMatteo Collina1h ago · 10 min readsqlite-js-stream: Litestream, rewritten for NodeSQLite is the most widely used database in the world, but it only runs on a single machine. If that machine fails, your database is lost. Unlike Postgres, you can't solve this with traditional replica1blog.platformatic.dev
PPowerduck9h ago · 4 min readYour AI Agent's Exit Code Is Lying to YouLast week I watched an agent finish a task with a clean exit 0, a green test suite, and a confident Done. in the chat. Nothing in the system had changed. The migration script it claimed to have run ha11Ablogs.powerduck.com
Zzifeng peng6h ago · 1 min readBuild from a shared goalThree coding agents on one checkout still drift unless the work converges on the same project thread. In practice that means: discuss the brief together, divide coding tasks around a common goal, run brainstormhq.hashnode.dev
VVin Patel9h ago · 2 min readAsk HN: How Do You Ship AI-Written Code You Don't Understand?Originally published at vinpatel.com How do you ship AI-written code you don't understand? You stop trying to understand it line by line, and you start owning what it does. That question is the title vinpatel.hashnode.dev
PPowerduck9h ago · 4 min readThe Agent Fixed the TypeScript Error by Adding `as any`The build was red. A type error in the response mapper: Property 'owner' does not exist on type 'TeamSummary'. The agent opened the file, stared at the screen for maybe a second, and rewrote the mappeblogs.powerduck.com
FFlutter Sensei12h ago · 10 min readHow to Parse AI-Generated JSON into a Dart List – A Step-by-Step E-Commerce GuideHey guys, welcome! When you're building a sleek e-commerce UI in Flutter, one of the first things you always need is realistic mock data. Sure, you could search online for a free dummy JSON API, but afluttersensei.hashnode.dev
PParthasarathy G1d ago · 24 min readThe best AI coding agent harness in 2026: five compared on one modelModel capability is only half of what you get from a coding agent. The other half is the harness: the loop, the system prompt, the tools, the permission rules, how it remembers things, how it shrinks agentnotes.hashnode.dev
Zzifeng peng1d ago · 1 min readSeparate the checks from the draftDraft and verify are different jobs. If the same agent writes the patch and grades it, you get confident wrong code — polished, fluent, and wrong in the same places the draft model always misses. Whatbrainstormhq.hashnode.dev
Zzifeng peng1d ago · 1 min readBuild from a shared goal before anyone writes codeDifferent AI minds only help when they share one brief. I keep seeing multi-agent setups that look busy and still fork the project: one agent invents a direction, another "reviews" in the same blind sbrainstormhq.hashnode.dev
Zzifeng peng1d ago · 1 min readKeep the next revision in the same projectThe first useful AI draft is rarely the last version you want to ship. What usually breaks the loop is the hand-off: the discussion lives in one chat, the build in another, and the second change requebrainstormhq.hashnode.dev
