Lanh Hoang lanhhoang

LLM Wiki v2

A pattern for building personal knowledge bases using LLMs. Extended with lessons from building agentmemory, a persistent memory engine for AI coding agents.

This builds on Andrej Karpathy's original LLM Wiki idea file. Everything in the original still applies. This document adds what we learned running the pattern in production: what breaks at scale, what's missing, and what separates a wiki that stays useful from one that rots.

What the original gets right

The core insight is correct: stop re-deriving, start compiling. RAG retrieves and forgets. A wiki accumulates and compounds. The three-layer architecture (raw sources, wiki, schema) works. The operations (ingest, query, lint) cover the basics. If you haven't read the original, start there.

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.

Linking a custom domain to a deployed application is confusing and at times you don't get a route on how to link the domain with the website. 😫

In this article, I will be explaining how to link a Custom domain brought in Google Domains service to an application hosted on Vercel.

Perks of using vercel is that it allows adding domain without any verification unlike Heroku/Netlify, which means you don't have to add your billing details to add a domain to your application. Moreover, it is fast, efficient, and easy to use.😁

What is Vercel?

Vercel is a deployment and collaboration platform for frontend developers. Vercel enables developers to host websites and web services that deploy instantly and scale automatically – all without any configuration. Source - Official Docs

Installing Fish shell on MacOS (Intel and M1) using brew

Fish is a smart and user-friendly command line (like bash or zsh). This is how you can instal Fish on MacOS and make your default shell.

Note that you need the https://brew.sh/ package manager installed on your machine.

Install Fish

brew install fish

	export const gpt_functions_param_jobs_fetching = [
	{
	name: "process_job_data",
	description: "Process job data and extract core fields for a scraper.",
	parameters: {
	type: "object",
	additionalProperties: false,
	properties: {
	// Company
	company_name: {

	from bs4 import BeautifulSoup
	import requests
	import re
	import urllib2
	import os
	import argparse
	import sys
	import json

	# adapted from http://stackoverflow.com/questions/20716842/python-download-images-from-google-image-search

	###
	###
	### UPDATE: For Win 11, I recommend using this tool in place of this script:
	### https://christitus.com/windows-tool/
	### https://github.com/ChrisTitusTech/winutil
	### https://www.youtube.com/watch?v=6UQZ5oQg8XA
	### iwr -useb https://christitus.com/win \| iex
	###
	### OR take a look at
	### https://github.com/HotCakeX/Harden-Windows-Security

	require 'date'
	require 'benchmark'

	n = 1_000_000
	start_date = Date.new(2012, 01, 01)
	end_date = Date.new(2012, 03, 01)
	act_date = Date.new(2012, 02, 01)

	Benchmark.bm(10) do \|x\|
	x.report('include?') do