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.

bobcallaway

# used in DevConf 22 presentation: https://www.youtube.com/watch?v=vh26wcpA1-M

#!/bin/sh
#set -X
#doitlive commentecho: true
# 
# Start with a clean slate 
make clean 
# 
# Edit the source

reegnz

Inspecting Kubernetes JWT tokens

Start a cluster with a dummy workload

kind create cluster
kubectl apply -f cli.yaml
kubectl apply -f discovery.yaml

ometa

// Golang example that creates an http client that leverages a SOCKS5 proxy and a DialContext
func NewClientFromEnv() (*http.Client, error) {
	proxyHost := os.Getenv("PROXY_HOST")

	baseDialer := &net.Dialer{
		Timeout:   30 * time.Second,
		KeepAlive: 30 * time.Second,
	}
	var dialContext DialContext