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.

jonstuebe

class ReSocket {
  private token: string | undefined;
  private socket: WebSocket | undefined;
  private listeners: boolean = false;
  private currentAttempt: number = 0;
  private backoffTime: number = 1000;
  private maxAttempts: number = 30;
  private timer: NodeJS.Timeout | undefined;
  private messageFn: (data: any) => void = (data) => {
    //

nnarhinen

/**
 * Module dependencies.
 */

var express = require('express'),
    http = require('http'),
    RedisStore = require('connect-redis')(express),
    redis = require('redis'),
    cookie = require('cookie'),
    connect = require('connect'),

SlexAxton

gifify() {
  if [[ -n "$1" ]]; then
    if [[ $2 == '--good' ]]; then
      ffmpeg -i $1 -r 10 -vcodec png out-static-%05d.png
      time convert -verbose +dither -layers Optimize -resize 600x600\> out-static*.png  GIF:- | gifsicle --colors 128 --delay=5 --loop --optimize=3 --multifile - > $1.gif
      rm out-static*.png
    else
      ffmpeg -i $1 -s 600x400 -pix_fmt rgb24 -r 10 -f gif - | gifsicle --optimize=3 --delay=3 > $1.gif
    fi
  else

max-mapper

// get latest canary and turn on web audio in chrome://flags

navigator.webkitGetUserMedia({audio:true}, function (stream) {
  var context = new webkitAudioContext()
  var streamSource = context.createMediaStreamSource(stream)
  var sent = false

  bufferFiller = context.createJavaScriptNode(2048, 1, 1)
  bufferFiller.onaudioprocess = function(e) {
    if (sent) return

piscisaureus

Locate the section for your github remote in the .git/config file. It looks like this:

[remote "origin"]
	fetch = +refs/heads/*:refs/remotes/origin/*
	url = git@github.com:joyent/node.git

Now add the line fetch = +refs/pull/*/head:refs/remotes/origin/pr/* to this section. Obviously, change the github url to match your project's URL. It ends up looking like this:

jlong

var parser = document.createElement('a');
parser.href = "http://example.com:3000/pathname/?search=test#hash";

parser.protocol; // => "http:"
parser.hostname; // => "example.com"
parser.port;     // => "3000"
parser.pathname; // => "/pathname/"
parser.search;   // => "?search=test"
parser.hash;     // => "#hash"
parser.host;     // => "example.com:3000"