<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Open Source on Farid</title><link>https://polymatx.dev/blog/tags/open-source/</link><description>Recent content in Open Source on Farid</description><generator>Hugo</generator><language>en</language><lastBuildDate>Tue, 04 Aug 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://polymatx.dev/blog/tags/open-source/index.xml" rel="self" type="application/rss+xml"/><item><title>Why Parallel Coding Agents Need a Coordination Layer (So I Built Dibs)</title><link>https://polymatx.dev/blog/posts/dibs-coordination-for-parallel-coding-agents/</link><pubDate>Tue, 04 Aug 2026 00:00:00 +0000</pubDate><guid>https://polymatx.dev/blog/posts/dibs-coordination-for-parallel-coding-agents/</guid><description>&lt;p&gt;Running one AI coding agent is last year&amp;rsquo;s workflow. My normal setup today is two or three Claude Code sessions working the same repository from separate git worktrees — one refactoring, one writing tests, one chasing a bug. It is a genuine productivity multiplier, with one recurring failure mode:&lt;/p&gt;
&lt;blockquote&gt;
&lt;p&gt;Agent A is halfway through refactoring &lt;code&gt;src/auth&lt;/code&gt;. Agent B decides the quickest fix for its bug also lives in &lt;code&gt;src/auth/token.go&lt;/code&gt;. Neither knows about the other. The result is silent overwrites or an unmergeable diff — produced by two tools that were each doing their job correctly.&lt;/p&gt;</description></item><item><title>GoFrame: A Batteries-Included Go Web Framework I Built for Real Projects</title><link>https://polymatx.dev/blog/posts/goframe-batteries-included-go-web-framework/</link><pubDate>Mon, 06 Jul 2026 00:00:00 +0300</pubDate><guid>https://polymatx.dev/blog/posts/goframe-batteries-included-go-web-framework/</guid><description>&lt;p&gt;Every Go service I&amp;rsquo;ve built over the years started the same way: pull in a router, wire up GORM, write the Redis connection helper &lt;em&gt;again&lt;/em&gt;, copy the RabbitMQ consumer boilerplate from the last project, glue a JWT middleware together, and spend the first two days of a new project rebuilding the same plumbing I built last time.&lt;/p&gt;
&lt;p&gt;Go&amp;rsquo;s standard library and ecosystem are excellent — that&amp;rsquo;s not the problem. The problem is that &amp;ldquo;lightweight and composable&amp;rdquo; means every team composes a slightly different stack, and the glue code between the pieces is where the bugs and the boredom live.&lt;/p&gt;</description></item><item><title>Weave: I Built a TypeScript-Native Agent Orchestrator (And Why You Might Want One)</title><link>https://polymatx.dev/blog/posts/weave-typescript-agent-orchestrator/</link><pubDate>Sun, 05 Jul 2026 00:00:00 +0000</pubDate><guid>https://polymatx.dev/blog/posts/weave-typescript-agent-orchestrator/</guid><description>&lt;p&gt;After months of building LLM agent pipelines for real projects, I kept hitting the same three walls: my graph state degraded into &lt;code&gt;any&lt;/code&gt; soup, wiring MCP servers into agents meant writing the same glue code over and over, and answering &amp;ldquo;why did this run cost $4?&amp;rdquo; required either a SaaS observability subscription or grepping JSON logs at 3 a.m.&lt;/p&gt;
&lt;p&gt;So I built &lt;a href="https://github.com/polymatx/weave" class="external-link" target="_blank" rel="noopener"&gt;weave&lt;/a&gt; — a small, opinionated framework for multi-agent workflows in TypeScript. It&amp;rsquo;s MIT licensed, published on npm, and the core is about 600 lines of code. This post is a tour of what it does and why I made the choices I made.&lt;/p&gt;</description></item></channel></rss>