ry

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.

pimeys

State of Async WASI in Rust

Let me share what I've learned about implementing async WASIp2 components in Rust. My goal is to get the entire Tokio ecosystem working together seamlessly. This isn't a complete test of the ecosystem - some things might be simpler than we expect. Check out dicej's wasi-socket-tests repository for examples.

The first obstacle: you'll need a nightly version of Rust. Without it, you'll need major ecosystem changes to avoid the wasip2 module in the Rust standard library and use wasi crates for the necessary functionality.

Let's walk through the steps to get Reqwest working with Tokio.

Socket2

veekaybee

ChatGPT Resources

Context

ChatGPT appeared like an explosion on all my social media timelines in early December 2022. While I keep up with machine learning as an industry, I wasn't focused so much on this particular corner, and all the screenshots seemed like they came out of nowhere. What was this model? How did the chat prompting work? What was the context of OpenAI doing this work and collecting my prompts for training data?

I decided to do a quick investigation. Here's all the information I've found so far. I'm aggregating and synthesizing it as I go, so it's currently changing pretty frequently.

Model Architecture

zxkane

#!/bin/bash

set -euxo pipefail


BUCKET=aws-gcr-solutions-assets

mirror-helm() {
    local repoName=$1
    local chartName=$2

jdolitsky

#!/usr/bin/env bash

set -o errexit
set -o nounset
set -o pipefail

if ! crane version >/dev/null; then echo "Must install crane."; exit 1; fi
if ! cosign version >/dev/null; then echo "Must install cosign."; exit 1; fi

GHCR_ROOT_NAMESPACE="${GHCR_ROOT_NAMESPACE:-}"

bcnzer

addEventListener('fetch', event => {
  event.respondWith(handleRequest(event))
})

/**
 * Entry point of the worker
 */
async function handleRequest(event) {
  try {
    // Get the JWT

jbeda

# These filters are based on email that Tim Hockin sent to kubernetes-dev a
# while ago on how to bring some sanity to github notifications with Kubernetes.
#
# https://groups.google.com/forum/#!msg/kubernetes-dev/5qU8irU7_tE/aZov0LpCBwAJ
#
# This file can be processed with https://github.com/mesozoic/gmail-yaml-filters

# Mark all k8s email
- match:
    any:

Ariex

const Page = require('puppeteer/lib/Page');

// the following 2 methods are originally from: https://github.com/GoogleChrome/puppeteer/issues/85#issuecomment-341743259, with some modification to fit puppeteerv1.0.0
async function newPageWithNewContext(browser) {
    const { browserContextId } = await browser._connection.send('Target.createBrowserContext');
    const { targetId } = await browser._connection.send('Target.createTarget', { url: 'about:blank', browserContextId });
    const target = await browser._targets.get(targetId);
    const client = await browser._connection.createSession(targetId);
    const page = await Page.create(client, target, browser._ignoreHTTPSErrors, browser._appMode, browser._screenshotTaskQueue);
    page.browserContextId = browserContextId;

3n21c0

package main

import (
	"context"
	"flag"
	"fmt"
	"log"
	"net/http"
	"os"
	"os/signal"

danielhavir

### Removes Node and NPM from mac OS ###
# Uninstall using NPM
npm uninstall npm -g || echo "NPM uninstall didn't work"
# Try if Node's installed with Brew
(brew uninstall node&&brew prune)||echo "Node was not installed with Brew"
# Remove all files and folders including Node or NPM
rm -rf /usr/local/{lib/node{,/.npm,_modules},bin,share/man}/{npm*,node*,man1/node*}
rm -rf /usr/local/bin/npm /usr/local/share/man/man1/node* /usr/local/lib/dtrace/node.d ~/.npm ~/.node-gyp /opt/local/bin/node opt/local/include/node /opt/local/lib/node_modules
rm -rf /usr/local/{lib/node{,/.npm,_modules},bin,share/man}/npm*
rm -rf /usr/local/include/node*