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.

bgulla

WLED Hyperion Ambilight Notes

Disclaimer

Nothing about this is turnkey. It's a mess of protocols, firmware and apps, but if you're the right kind of nuts, keep reading.

Required hardware

• Power Supply Amazon $22
• WS2812B Addressable RGB LED Strips Amazon $30
• NodeMCU 8266 Amazon $6
• AndroidTV - There are alternatives that include using a HDMI grabber and decoding the signal on a Pi/server directly, but this works for a 0.1 alpha attempt. I used a NVidia ShieldTV

danieljs777

package PasswordEncryption;

/**
 *
 * @author daniel
 */
import java.io.Console;
import java.io.IOException;
import java.security.*;
import java.io.ByteArrayOutputStream;

david-wm-sanders

.\venv\Scripts\pip freeze | Out-File -Encoding UTF8 requirements.txt

pascalandy

---
layout: default
permalink: /
---

# Advanced JavaScript syllabus

***If you are a teacher or interested in the design of the course, see the [meta](https://github.com/advanced-js/syllabus/blob/master/meta.md) document.***

* **Course:** [INFO1-CE9766, NYU SCPS](http://scps.nyu.edu/content/scps/academics/course_detail.html?id=INFO1-CE9766)

mrklein

%matplotlib inline

filename = 'vtk-plot_0.vtk'

import vtk
from numpy import zeros
import matplotlib.pyplot as plt

reader = vtk.vtkUnstructuredGridReader()
reader.SetFileName(filename)

PurpleBooth

Project Title

One Paragraph of project description goes here

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.

Prerequisites

tdd

Workbench

We extract our test repos from this small Zip file.

• main is a "container" repository with its working copy,
• plugin is a "shared" repository with its working copy,
• remotes emulates remote bare repos for both, to better resemble regular usage.

The idea is to use plugin as a subtree of main in a vendor/plugins/demo path, and allow maintenance both ways:

vlucas

#!/bin/bash

# @link https://gist.github.com/mattscilipoti/8424018
#
# Called by "git push" after it has checked the remote status,
# but before anything has been pushed.
#
# If this script exits with a non-zero status nothing will be pushed.
#
# Steps to install, from the root directory of your repo...

nateware

# IE is still braindead so still use favicon.ico

convert -resize x16 -gravity center -crop 16x16+0+0 -flatten -colors 256 input.png output-16x16.ico
convert -resize x32 -gravity center -crop 32x32+0+0 -flatten -colors 256 input.png output-32x32.ico
convert output-16x16.ico output-32x32.ico favicon.ico

# Then, HTML needs to specify size="XxY" as largest size due to browser bugs

<link rel="shortcut icon" href="/favicon.ico" sizes="32x32">