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.

cassidoo

Talk to Me page

Who am I?

...

What do I do at organization?

...

wietze

import requests

##########
### MITRE ATT&CK ONELINERS
###   for constructing Python objects 
###   with all ATT&CK techniques in them
###   using the latest MITRE ATT&CK data
##########

# Get MITRE ATT&CK technique objects as list

klaaspieter

From time to time, Musk will send out an e-mail to the entire company to enforce a new policy or let them know about something that's bothering him. One of the more famous e-mails arrived in May 2010 with the subject line: Acronyms Seriously Suck:

There is a creeping tendency to use made up acronyms at SpaceX. Excessive use of made up acronyms is a significant impediment to communication and keeping communication good as we grow is incredibly important. Individually, a few acronyms here and there may not seem so bad, but if a thousand people are making these up, over time the result will be a huge glossary that we have to issue to new employees. No one can actually remember all these acronyms and people don't want to seem dumb in a meeting, so they just sit there in ignorance. This is particularly tough on new employees.

That needs to stop immediately or I will take drastic action - I have given enough warning over the years. Unless an acronym is approved by me, it should not enter the SpaceX glossary.

marcelom

#!/usr/bin/env python

## Tiny Syslog Server in Python.
##
## This is a tiny syslog server that is able to receive UDP based syslog
## entries on a specified port and save them to a file.
## That's it... it does nothing else...
## There are a few configuration parameters.

LOG_FILE = 'youlogfile.log'