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.

georgeh

sensor:
    # Read in DTE Insight via MQTT. You can add an MQTT integration in Home Assistant with the DTE Insight's IP address on port 2883
  - platform: mqtt # this is my DTE Insight, if you have multiple MQTT servers this might be different
    name: Instant Energy Usage
    state_topic: "event/metering/instantaneous_demand"
    unit_of_measurement: W
    value_template: "{{ value_json.demand }}"
    state_class: measurement
    device_class: energy

brandond

#!/bin/bash

FLAGS=""
SOURCE="docker.io/library"
DEST="docker.io/brandond"
IMAGE="busybox"
TAG="1.32.0-uclibc"

docker buildx imagetools inspect ${SOURCE}/${IMAGE}:${TAG} --raw | \
  jq -r '.manifests[] | (.digest + " " + .platform.architecture)' | \

AlinaNova21

Kubernetes and Arm

Getting rke and Rancher setup to run kubernetes on arm is interesting. There is no official support yet via rancher, although there is interest and some work done towards those efforts. This is my attempt at getting a cluster of 3 Pis (2 3Bs and 1 3B+) provisioned and registered to a rancher 2 server.

Prep

I've successfully completed this both with Hypriot OS 1.9.0 and the arm64 builds https://github.com/DieterReuter/image-builder-rpi64 Both times I used the same basic cloud-init setup

igrigorik

def tip(msg); puts; puts msg; puts "-"*100; end

#
# 30 Ruby 1.9 Tips, Tricks & Features:
# http://www.igvita.com/2011/02/03/new-ruby-19-features-tips-tricks/
#

tip "Upgrading to Ruby 1.9 is simple: rvm install 1.9.2 && rvm --default 1.9.2"

tip "Ruby 1.9 supports named captures in regular expressions!"

lucasefe

if __FILE__ == $0
  puts "Run with: watchr #{__FILE__}. \n\nRequired gems: watchr rev"
  exit 1
end

# --------------------------------------------------
# Convenience Methods
# --------------------------------------------------
def run(cmd)
  puts(cmd)