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.

jonasporto

setting autocorrect from git commands

git config --global help.autocorrect 1

ex:

git pulll origin master

WARNING: You called a Git command named 'pulll', which does not exist. Continuing under the assumption that you meant 'pull' in 0.1 seconds automatically...

alexcorvi

function mime2ext($mime){
  $all_mimes = '{"png":["image\/png","image\/x-png"],"bmp":["image\/bmp","image\/x-bmp","image\/x-bitmap","image\/x-xbitmap","image\/x-win-bitmap","image\/x-windows-bmp","image\/ms-bmp","image\/x-ms-bmp","application\/bmp","application\/x-bmp","application\/x-win-bitmap"],"gif":["image\/gif"],"jpeg":["image\/jpeg","image\/pjpeg"],"xspf":["application\/xspf+xml"],"vlc":["application\/videolan"],"wmv":["video\/x-ms-wmv","video\/x-ms-asf"],"au":["audio\/x-au"],"ac3":["audio\/ac3"],"flac":["audio\/x-flac"],"ogg":["audio\/ogg","video\/ogg","application\/ogg"],"kmz":["application\/vnd.google-earth.kmz"],"kml":["application\/vnd.google-earth.kml+xml"],"rtx":["text\/richtext"],"rtf":["text\/rtf"],"jar":["application\/java-archive","application\/x-java-application","application\/x-jar"],"zip":["application\/x-zip","application\/zip","application\/x-zip-compressed","application\/s-compressed","multipart\/x-zip"],"7zip":["application\/x-compressed"],"xml":["application\/xml","text\/xml"],"svg":

vitorbritto

Remove MySQL completely

1. Open the Terminal

2. Use mysqldump to backup your databases

3. Check for MySQL processes with: ps -ax | grep mysql

4. Stop and kill any MySQL processes

5. Analyze MySQL on HomeBrew:

   brew remove mysql
   

octocat

# Compiled source #
###################
*.com
*.class
*.dll
*.exe
*.o
*.so

# Packages #

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.

ruanltbg

Apache

$ sudo apt-get update
$ sudo apt-get install apache2
# activate mod_rewrite
$ sudo a2enmod rewrite
#restart
$ sudo service apache2 restart

Mysql

timruffles

module HashConverter
  class << self
    def to_underscore hash
      convert hash, :underscore
    end
    def to_camel_case hash
      convert hash, :camelize, :lower
    end
    def convert obj, *method
      case obj