Gihan Karunarathne gihankarunarathne

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.

Brew Bundle Brewfile Tips

Copyright & License

Unless otherwise noted (either in this file or in a file's copyright section) the contents of this gist are Copyright ©️2020 by Christopher Allen, and are shared under spdx:Creative Commons Attribution Share Alike 4.0 International (CC-BY-SA-4.) open-source license.

Sponsor

If you more tips and advice like these, you can become a monthly patron on my GitHub Sponsor Page for as little as $5 a month; and your contributions will be multipled, as GitHub is matching the first $5,000! This gist is all about Homebrew, so if you like it you can support it by donating to them or becoming one of their Github Sponsors.

	"""
	The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
	This file is the complete algorithm.
	Everything else is just efficiency.

	@karpathy
	"""

	import os # os.path.exists
	import math # math.log, math.exp

	def nullTrustManager = [
	checkClientTrusted: { chain, authType -> },
	checkServerTrusted: { chain, authType -> },
	getAcceptedIssuers: { null }
	]

	def nullHostnameVerifier = [
	verify: { hostname, session ->
	//true
	hostname.startsWith('yuml.me')

	private static WritableMap convertJsonToMap(JSONObject jsonObject) throws JSONException {
	WritableMap map = new WritableNativeMap();

	Iterator<String> iterator = jsonObject.keys();
	while (iterator.hasNext()) {
	String key = iterator.next();
	Object value = jsonObject.get(key);
	if (value instanceof JSONObject) {
	map.putMap(key, convertJsonToMap((JSONObject) value));
	} else if (value instanceof JSONArray) {

	ssh-keygen -t rsa -b 4096 -m PEM -f jwtRS256.key
	# Don't add passphrase
	openssl rsa -in jwtRS256.key -pubout -outform PEM -out jwtRS256.key.pub
	cat jwtRS256.key
	cat jwtRS256.key.pub

	function mapValues(obj, fn) {
	return Object.keys(obj).reduce((result, key) => {
	result[key] = fn(obj[key], key);
	return result;
	}, {});
	}

	function pick(obj, fn) {
	return Object.keys(obj).reduce((result, key) => {
	if (fn(obj[key])) {

	minikube start --kubernetes-version=v1.7.0 --extra-config=apiserver.Authorization.Mode=RBAC

	kubectl create clusterrolebinding add-on-cluster-admin --clusterrole=cluster-admin --serviceaccount=kube-system:default

	minikube dashboard

	#include <unistd.h>
	#include <node.h>
	#include <string.h>
	#include <v8.h>

	using namespace v8;

	unsigned long long count = 0;

	// native blocking/compute intensive function

	import requests

	#http://docs.python-requests.org/en/latest/user/quickstart/#post-a-multipart-encoded-file

	url = "http://localhost:5000/"
	fin = open('simple_table.pdf', 'rb')
	files = {'file': fin}
	try:
	r = requests.post(url, files=files)
	print r.text