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.

pratos

# For Windows users# Note: <> denotes changes to be made

#Create a conda environment
conda create --name <environment-name> python=<version:2.7/3.5>

#To create a requirements.txt file:
conda list #Gives you list of packages used for the environment

conda list -e > requirements.txt #Save all the info about packages to your folder

adammacias

GIT - Primeiros passos

Instalação e configurações básicas

1. Baixe e instale o Git

==================

2. Configurar chave de segunça

rdeavila

#!/bin/bash
# Adicione um novo remote; pode chamá-lo de "upstream":

git remote add upstream https://github.com/usuario/projeto.git

# Obtenha todos os branches deste novo remote,
# como o upstream/master por exemplo:

git fetch upstream