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.

lmas

// Copyright © 2016 Alex
//
// This program is free software: you can redistribute it and/or modify it under
// the terms of the GNU Affero General Public License as published by the Free
// Software Foundation, either version 3 of the License, or (at your option) any
// later version.
//
// This program is distributed in the hope that it will be useful, but WITHOUT
// ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
// FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more

justincase

type T struct {
    A int
    B string
}

t := T{23, "skidoo"}
s := reflect.ValueOf(&t).Elem()
typeOfT := s.Type()

for i := 0; i < s.NumField(); i++ {

witscher

description     "Unicorn"

# starting unicorn with bundler, and a user contained rvm:


start on filesystem or runlevel [2345]
stop on runlevel [!2345]


chdir /path/to/current

sasha-id

MongoDB upstart scripts for Ubuntu.

Run following commands after installing upstart scripts:

ln -s /lib/init/upstart-job /etc/init.d/mongoconf
ln -s /lib/init/upstart-job /etc/init.d/mongodb
ln -s /lib/init/upstart-job /etc/init.d/mongos

To start services use:

robertjwhitney

# This function cleans up messy HTML that was pasted by a user to a WYSIWYG editor.
# Specifically it also handles messy Word\Outlook generated HTML while keeping its original formattings.

require 'rubygems'
require 'sanitize'

  def clean_up_document(html)
    elements = %w[p b h1 h2 h3 h4 h5 h6 strong li ul ol i br div pre p]
    attributes = {
                  'a' => ['href', 'title'], 

mxcl

#!/bin/sh
# Just copy and paste the lines below (all at once, it won't work line by line!)
# MAKE SURE YOU ARE HAPPY WITH WHAT IT DOES FIRST! THERE IS NO WARRANTY!

function abort {
  echo "$1"
  exit 1
}

set -e

boj

class Score
  include MongoMapper::Document
  
  key :user_id, ObjectId,   :index => true
  key :stage,   String,     :index => true
  key :score,   Integer
  timestamps!
  
  def self.high_score_map
    <<-MAP

chischaschos

*.log

lajunta

#!/bin/sh 
iptables -F
iptables -X
iptables -P INPUT DROP
iptables -P FORWARD DROP
iptables -P OUTPUT DROP
iptables -t nat -F
iptables -t nat -P PREROUTING ACCEPT
iptables -t nat -P POSTROUTING ACCEPT
iptables -t nat -P OUTPUT ACCEPT