Skip to content

Instantly share code, notes, and snippets.

View csnyder256's full-sized avatar

Cade csnyder256

3 followers · 2 following
View GitHub Profile
@csnyder256
csnyder256 / PLATFORM-SNAPSHOT.md
Created July 22, 2026 05:01
RAG-OS in practice: an anonymized reference build snapshot

RAG-OS in practice: a reference build snapshot

About this document

This is a point-in-time, anonymized snapshot of one working build of the architecture described in BUILD-GUIDE.md. It is condensed from the maintainer's internal, code-verified platform reference, which documents the system as the code on disk actually behaves and flags where older prose disagrees with the code.

It is scrubbed of anything machine-specific or personal: names, absolute paths, private repository names, credentials, and the identifying specifics of a separate high-stakes always-on workload that shares the same machine (in this build, an automated trading system). Module and file names under system/ are kept, because they are the useful part of a reference.

Read it as a worked example, not a specification. The build guide asks you a question at every design fork, so a system you build from it will differ. This snapshot shows what one set of answers looks like after the system has been running for a wh

@csnyder256
csnyder256 / LLM-RAG-OS.md
Last active July 22, 2026 14:05
A 2nd brain OS for any LLM, fully genericized architecture spec with explicit builder fork points. Remote control, RAG, multi-agent/cross-family dispatching, sharing the same CLI that avoids degradation while continuously improving cross-repo/cross-org knowledge. Share this file with any agent.

Build Your Own Personal AI Operating System

A brand-, harness-, and frontend-agnostic blueprint you hand to a coding agent.

This document describes how to build a persistent, always-on personal AI system: a supervised background process that turns an incoming message into either an answer or real work in a real code repository, accumulates knowledge that compounds over time, runs governed background jobs, stays reachable from your phone or terminal, and never lets a language model's context window, an untrusted document, or a compromised account cross a safety boundary.

It is not a chatbot. The intelligent layer decides and delegates. It does not hold the durable state itself. The founding rule that everything else follows from: durable state lives in files and a database, brains are ephemeral, and the coordinating kernel is small and boring. No model context outlives a single session. Anything that must survive a crash, a restart, or a model downgrade lives on disk, never in a context window.