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.

equivalent

• Rails 7.0.2
• stimulus js
• importmaps
• dropzone.js
• direct upload ActiveStorage

This is simple implementation of technologies in hobby project of mine built in Rails7 where I need direct upload to S3.

kottenator

// Implementation in ES6
function pagination(c, m) {
    var current = c,
        last = m,
        delta = 2,
        left = current - delta,
        right = current + delta + 1,
        range = [],
        rangeWithDots = [],
        l;

denji

Moved to git repository: https://github.com/denji/nginx-tuning

NGINX Tuning For Best Performance

For this configuration you can use web server you like, i decided, because i work mostly with it to use nginx.

Generally, properly configured nginx can handle up to 400K to 500K requests per second (clustered), most what i saw is 50K to 80K (non-clustered) requests per second and 30% CPU load, course, this was 2 x Intel Xeon with HyperThreading enabled, but it can work without problem on slower machines.

You must understand that this config is used in testing environment and not in production so you will need to find a way to implement most of those features best possible for your servers.

nightire

Routes

小心地使用 Match（Rails 3 已实现）

Rails 3 提供了 match 方法供我们自定义 routes，然而我们要小心使用它以避免“跨站脚本攻击”（XSS Attack）。比如像这样的 routes：

注：（r3 代表 Rails 3，r4 代表 Rails 4）

# routes.rb

losingle

Ruby Debugger

Most of the time simple output statements using warn, raise, or a logger will help you find your issue. But sometimes you need the big guns, and that means ruby-debug.

Ruby-Debug

The ruby-debug package has had some rocky times during the transition from Ruby 1.8.7 to 1.9.2 and beyond. But now, finally, the debugger is reliable and usable.

Installation

kamens

function KnowledgeMapInitGlobals() {

    window.KnowledgeMapGlobals = {

        colors: {
            blue: "#0080C9",
            green: "#8EBE4F",
            red: "#E35D04",
            gray: "#FFFFFF"
        },

jaigouk

require 'amqp'
module HiringThingEM
  def self.start
    if defined?(PhusionPassenger)
      PhusionPassenger.on_event(:starting_worker_process) do |forked|
      # for passenger, we need to avoid orphaned threads
        if forked && EM.reactor_running?
          EM.stop
        end
        Thread.new {

abarringer

#!/usr/bin/env ruby

require 'logger'
require 'rest_client'

$stdout.sync = true
$stdin.sync = true

path = "/usr/local/var/log/ejabberd/auth.log"
file = File.open(path, File::WRONLY | File::APPEND | File::CREAT)

karmi

# Run me with:
#
#     $ nginx -p /path/to/this/file/ -c nginx.conf
#
# All requests are then routed to authenticated user's index, so
#
#    GET http://user:password@localhost:8080/_search?q=*
#
# is rewritten to:
#