larrynung

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.

larrynung

Beast Mode

Beast Mode is a custom chat mode for VS Code agent that adds an opinionated workflow to the agent, including use of a todo list, extensive internet research capabilities, planning, tool usage instructions and more. Designed to be used with 4.1, although it will work with any model.

Below you will find the Beast Mode prompt in various versions - starting with the most recent - 3.1

Installation Instructions

• Go to the "agent" dropdown in VS Code chat sidebar and select "Configure Modes".
• Select "Create new custom chat mode file"

larrynung

Latency Comparison Numbers (~2012)
----------------------------------
L1 cache reference                           0.5 ns
Branch mispredict                            5   ns
L2 cache reference                           7   ns                      14x L1 cache
Mutex lock/unlock                           25   ns
Main memory reference                      100   ns                      20x L2 cache, 200x L1 cache
Compress 1K bytes with Zippy             3,000   ns        3 us
Send 1K bytes over 1 Gbps network       10,000   ns       10 us
Read 4K randomly from SSD*             150,000   ns      150 us          ~1GB/sec SSD

larrynung

# Optimized my.cnf configuration for MySQL/MariaSQL on cPanel/WHM servers
#
# by Fotis Evangelou, developer of Engintron (engintron.com)
#
# === Updated December 2018 ===
#
# The settings provided below are a starting point for a 2GB - 4GB RAM server with 2-4 CPU cores.
# If you have less or more resources available you should adjust accordingly to save CPU,
# RAM and disk I/O usage.
# The settings marked with a specific comment or the word "UPD" after the value

larrynung

#!/usr/bin/perl
#
# mysqldump -routines --no-create-info --no-data --no-create-db --compact | perl $0

use strict;
use warnings;
use utf8;

my $output_dir = "out";

larrynung

#!/bin/bash
# Simple inspector for Docker container IP address
# https://gist.github.com/mugifly/3bb1c54b48764340eda8

print_help () {
    echo -e "USAGE: docker-container-ip.sh KEYWORD\n"
    echo "KEYWORD: A keyword string for target container."
    echo "    e.g. dokku.mysql.foobar"
    echo -e "\n[Example usecases]"
    echo "  Connect to MySQL database in container:"

larrynung

#!/bin/bash -eu
#
# This script can sync MySQL schema without `DROP TABLE`.
# And If table isn't exists, import tables from source.
#
# Finally there is a confirmation screen.
#
# Need these packages: mysql, mysqldump, mysqldiff
# GitHub: @rluisr

larrynung

Commit Message Conventions

These rules are adopted from the AngularJS commit conventions.

• Goals
• Generating CHANGELOG.md
  • Recognizing unimportant commits
  • Provide more information when browsing the history
• Format of the commit message
• Subject line

larrynung

using System;

namespace StackExchange.Redis
{
    static class RedisKeyspaceNotifications
    {
        /// <summary>
        /// NOTE: For this sample to work, you need to go to the Azure Portal and configure keyspace notifications with "Kxge$" to
        ///       1) turn on expiration notifications (x), 
        ///       2) general command notices (g) and 

larrynung

Download courses from learning sites with youtube-dl

You can download whole courses from an array of tutorial sites with the CLI tool youtube-dl. In the example further down I'm using my Pluralsight account to get videos from a course at their site. Here is a list of all supported sites that you can download from with this tool

The flags you have to supply may vary depending on which site you make a request to.

You can get a free 3 month trial to Pluralsight by signing up for free to Visual Studio Dev Essentials

Installation