karpathy

LLM Wiki

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

The core idea

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

jhintux

import reactRefresh from '@vitejs/plugin-react-refresh';
import {defineConfig} from 'vite';
import stdLibBrowser from 'node-stdlib-browser';
import inject from '@rollup/plugin-inject';

export default defineConfig({
  resolve: {
    alias: stdLibBrowser
  },
  optimizeDeps: {

lewiuberg

Source: https://gregology.net/2018/09/raspberry-pi-time-machine/

Raspberry Pi Time Machine

Setup Pi with Raspbian

Install Raspbian Buster Lite on the SD card. Follow the instructions on the Installing images tutorial on the Raspberry Pi site. Then enable SSH. Insert the SD card into the Pi, plug in the USB hard drive, plug in a network cable, and power on the Pi.

Open Preferences and Raspberr pi configuration. Change the hostname to: "AirPort"

Next set a static IP address for your Raspberry Pi. Depending on your setup, you can either set the Pi to have a manual ip address or have your router assign a static IP address.