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.

jacekkarczmarczyk

<template>
    <div :style="containerStyle" class="vertical-tabs" :class="{'vertical-tabs--vertical-text': verticalText, 'vertical-tabs--horizontal-text': !verticalText}">
        <v-tabs :value="value" @input="$emit('input', $event)" :show-arrows="verticalText" :color="color" :slider-color="sliderColor" :style="tabsStyle">
            <v-tab v-for="item in items" :key="item">{{ item }}</v-tab>
        </v-tabs>
    </div>
</template>

<script>
  export default {

kboek

using NAudio.Wave;

using System;
using System.Net.WebSockets;

namespace Raskenlund  
{
    class Recorder 
    {
        private WaveInEvent waveIn;

matthewjberger

1.) Download a Nerd Font

2.) Unzip and copy to ~/.fonts

3.) Run the command fc-cache -fv to manually rebuild the font cache

nfriedly

// Note: The official .net SDK is in progress. It doesn't support streaming Speech to Text at the time of writing, 
// but it will soon. Please check it out before using this code.
//
// https://github.com/watson-developer-cloud/dotnet-standard-sdk

using System;
using System.Net.WebSockets;
using System.Net;
using System.Runtime.Serialization.Json;
using System.Threading;

ronnieoverby

public static class SftpExtensions
{
    /// <remarks> 
    /// unfortunately the ssh library doesn't expose a modern async api
    /// that supports cancellation via CancellationToken
    /// so we have to wrap the APM pattern with a TaskCompletionSource
    /// </remarks>
    public static Task UploadFileAsync(this SftpClient sftpClient, Stream input, string path, bool canOverride, Action<ulong> uploadCallback = null, CancellationToken cancellationToken = default(CancellationToken))
    {
        var tcs = new TaskCompletionSource<SftpUploadAsyncResult>();

webdevwilson

import java.util.HashMap;
import java.util.Map;

public enum State {

    ALABAMA("Alabama", "AL"), ALASKA("Alaska", "AK"), AMERICAN_SAMOA("American Samoa", "AS"), ARIZONA("Arizona", "AZ"), ARKANSAS(
    "Arkansas", "AR"), CALIFORNIA("California", "CA"), COLORADO("Colorado", "CO"), CONNECTICUT("Connecticut", "CT"), DELAWARE(
    "Delaware", "DE"), DISTRICT_OF_COLUMBIA("District of Columbia", "DC"), FEDERATED_STATES_OF_MICRONESIA(
    "Federated States of Micronesia", "FM"), FLORIDA("Florida", "FL"), GEORGIA("Georgia", "GA"), GUAM("Guam", "GU"), HAWAII(
    "Hawaii", "HI"), IDAHO("Idaho", "ID"), ILLINOIS("Illinois", "IL"), INDIANA("Indiana", "IN"), IOWA("Iowa", "IA"), KANSAS(