Dan Wang abcde090

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.

AWS Certified Solutions Architect Professional Cheat Cheets for Senior Engineer

Practice

SLA and Multi-Regions setup

For 2 9s (99%) and 3 9s (99.9%) scenario, a single region setup is enough.

Primitive Types & Reference Types in JavaScript

An explanation of JavaScript's pass-by-value, which is unlike pass-by-reference from other languages.

Facts

JavaScript has 2 kinds of variable types: primitive and reference.
A fixed amount of memory is reserved after creation of every variable.
When a variable is copied, it's in-memory value is copied.
Passing a variable to a function via a call also creates a copy of that variable.

Primitive Types

MacOS X + oh my zsh + powerline fonts + visual studio code (vscode) terminal settings

Thank you everybody, Your comments makes it better

Install oh my zsh

http://ohmyz.sh/

sh -c "$(curl -fsSL https://raw.github.com/ohmyzsh/ohmyzsh/master/tools/install.sh)"

	######################################################################################
	## Automating Infrastructure on Google Cloud with Terraform: Challenge Lab # GSP345 ##
	######################################################################################

	====================== Setup : Create the configuration files ======================
	Make the empty files and directories in Cloud Shell or the Cloud Shell Editor.
	------------------------------------------------------------------------------------

	touch main.tf
	touch variables.tf

	image: <your_image>

	pipelines:
	default:
	- step:
	name: <description>
	deployment: <your_deployment_name>
	image: google/cloud-sdk:320.0.0-alpine
	script:
	# on linux you can use `$ base64 -w 0 < ~/.google_app_credentials.json` to generate $GCP_KEY_FILE

	#!/bin/bash
	set -uxo pipefail

	# DISK_NAME = Name of the disk in terraform
	# DEVICE_NAME = When $DISK_NAME is mounted in the compute instance at `/dev/`

	MOUNT_DIR=/mnt/disks/persistent_storage

	# Check if entry exists in fstab
	grep -q "$MOUNT_DIR" /etc/fstab

	/* -------------------------------------------------------------------------- */
	// All Bootstrap 4 Sass Mixins [Cheat sheet]
	// Updated to Bootstrap v4.5.x
	// @author https://anschaef.de
	// @see https://github.com/twbs/bootstrap/tree/master/scss/mixins
	/* -------------------------------------------------------------------------- */

	/*
	// ########################################################################## */
	// New cheat sheet for Bootstrap 5: