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.

kmplngj

# This ESPHome configuration file is for a GeekMagic Display device (esp01).
# It sets up the device to display pages of text and notifications,
# with customizable intervals and padding. The display is updated
# periodically and can show different pages based on Home Assistant sensors.

# # License & Terms
# **Free and Open Use** - This code may be freely used, modified, and distributed by anyone.
# **Copyleft Requirement** - Any derivative works, modifications, or forks must also be 
# distributed freely under the same terms. You cannot restrict access to modified versions.
# **Attribution Note** - This code was primarily written with AI assistance.

vfarcic

# Source: https://gist.github.com/0431989df4836eb82bdac0cc53c7f3d6

# Used in https://youtu.be/R6OeIgb7lUI

##############################
# Flux 2 With GitOps Toolkit #
##############################

# What Is GitOps And Why Do We Want It?: https://youtu.be/HKkhD6nokC8
# Argo CD: Applying GitOps Principles To Manage Production Environment In Kubernetes: https://youtu.be/vpWQeoaiRM4

debsahu

esphome:
  name: veneer_display
  platform: ESP8266
  board: d1_mini

wifi:
  networks:
  - ssid: !secret esphome_wifi_ssid1
    password: !secret esphome_wifi_pass1
  # - ssid: !secret esphome_wifi_ssid2

Paraphraser

Recipe: from MQTT to InfluxDB via Node-RED

• 2024-05-16 adds case study explaining how to extract values from topic strings, and how to specify the measurement name in the message payload provided to the "influxdb out" node.
• 2023-12-02 revise graphics to correspond with "influxdb out" node, and explain the pros and cons of InfluxDB 1.x vs InfluxDB 2.

Introduction

Getting data produced by IoT sensors into a database is practically a mandatory step before effective visualisation (eg dashboards).

This recipe shows you how to get an MQTT payload into an InfluxDB database using three Node-RED nodes. It makes the following assumptions:

spacehuhn

In the end of setup():

  pinMode(D6, INPUT_PULLUP); // enable button pin

At the end of loop():

  if(digitalRead(D6)){
    if(!attack.isRunning()){
      serialInterface.runCommand("stopap");                    // stop access point and web interface

AustinSaintAubin

/*===========================================================================
  Title: I2C Scanner
    - Repository: https://gist.github.com/AustinSaintAubin/dc8abb2d168f5f7c27d65bb4829ca870
  
  Version: 8
  Date: 2018 / 02 / 17
  Author: Austin St. Aubin
  Email: AustinSaintAubin@gmail.com
  
  Description: