Anurag shenoy-anurag

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.

This was a proper pain... Turns out there's a whole host of issues with the rtl8188eus driver.

Steps to fix (on Debian / RaspberryPi OS).

1. Install build tools

sudo apt install -y \
  build-essential \
 git \

Plotting

Last Update: May 13, 2019
Offline Version

	<?xml version="1.0" encoding="UTF-8"?>
	<OpenSearchDescription xmlns="http://a9.com/-/spec/opensearch/1.1/" xmlns:moz="http://www.mozilla.org/2006/browser/search/">
	<ShortName>PWB</ShortName>
	<Description>PayWallBuster Search</Description>
	<InputEncoding>UTF-8</InputEncoding>
	<Image width="16" height="16" type="image/x-icon">https://paywallbuster.com/favicon-32x32.png</Image>
	<Url type="text/html" method="GET" template="https://paywallbuster.com/articles?article={searchTerms}"/>
	<moz:SearchForm>https://paywallbuster.com/articles</moz:SearchForm>
	</OpenSearchDescription>

	"""
	Sometimes when you run `pip freeze > requirements.txt` when using a conda environment,
	it spits out the exact locations of the packages on the local machine.

	This script allows you to run `pip list` instead, and
	convert the output of `pip list` into the format used by `requirements.txt` files.
	"""

	import re

	#!/usr/bin/env bash

	alias ip="dig +short myip.opendns.com @resolver1.opendns.com" # returns public IP address.
	alias localip="ipconfig getifaddr en0" # returns private/local network DHCP IP address. change to en1 if using ethernet.

	service_name="mongod"
	sudo service $service_name status # shows status of service_name. eg: sudo service mongod status

	Please provide the AWS S3 Bucket Region: us-east-2
	Use AWS Signature Version 4?[Y/n] Y
	is_aws_signature_v4: True
	Testing Connection...
	[OK] Connection To S3 Account
	[OK] Write Permissions
	[ERROR] Read Permissions
	Error Connecting to S3: Please check you have read permissions on the S3 bucket.
	Details: Traceback (most recent call last):
	File "/Users/anurags/Projects/DataOps/diffgram/install.py", line 290, in validate_s3_connection

	apiVersion: eksctl.io/v1alpha5
	availabilityZones:
	- us-east-2b
	- us-east-2a
	- us-east-2c
	cloudWatch:
	clusterLogging: {}
	iam:
	vpcResourceControllerPolicy: true
	withOIDC: false

	console.log("Hello World");

	let id: number = 3.14;
	let num_name: string = "Pi";
	let isRational: boolean = true;
	let ch: any = 'c'
	let n = undefined
	let n2 = null
	console.log(typeof n)

	FROM python:3.9

	# https://code.visualstudio.com/docs/remote/containers-advanced#_creating-a-nonroot-user
	ARG USERNAME=keras-vscode
	ARG USER_UID=1000
	ARG USER_GID=$USER_UID

	# Create the user
	RUN groupadd --gid $USER_GID $USERNAME \
	&& useradd --uid $USER_UID --gid $USER_GID -m $USERNAME \

Anurag shenoy-anurag

LLM Wiki

The core idea

1. Install build tools

Plotting

Contents