Tom Bartek itsx

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.

Inside the Claude Code source

Anthropic's Claude Code CLI source code leaked onto GitHub recently. All of it. About 1,900 files and a lot of TypeScript.

I read through the key modules. What follows is a breakdown of the surprising parts: how the system actually works, where Anthropic made clever engineering choices, and where their approach diverges from OpenAI's Codex in ways you wouldn't guess from using either tool.

Lifecycle of a request

Here's what happens when you type a message into Claude Code:

Python relative imports in AWS Lambda fail with attempted relative import with no known parent package

The Problem

In AWS Lambda if I attempt an explicit relative import like this

.
├── lambda_file.py
└── example.py

Open python

python

and type

>>> import ssl
>>> ssl.OPENSSL_VERSION
'OpenSSL 1.0.1g 7 Apr 2014'

Overview

Some notes about:

Explaining why current day Linux memory swap thrashing still happens (as of 2016).
Mitigating "stop the world" type thrashing issues on a Linux workstation when it's under high memory pressure and where responsiveness is more important than process completion.
Prioritizing and limiting memory use.
Older ulimit versus newer CGroup options.

These notes assume some basic background knowledge about memory management, ulimits and cgroups.

	#!/bin/bash

	#
	# copy this into /etc/profile.d/instance-name-tag.sh
	#
	# you will need:
	# - curl, jq, and aws cli installed
	# - an IAM role that gives the EC2 instance access to describe tags
	#

	#!/bin/sh

	# Make sure only root can run our script
	if [ "$(id -u)" != "0" ]; then
	echo "This script must be run as root" 1>&2
	exit 1
	fi

	# Load configuration
	. /etc/route53/config

	# Redis Cheatsheet
	# All the commands you need to know


	redis-server /path/redis.conf # start redis with the related configuration file
	redis-cli # opens a redis prompt


	# Strings.

	#!/bin/bash
	# file: ttfb.sh
	# curl command to check the time to first byte
	# usage
	# 1. ./ttfb.sh "https://google.com"
	# 2. seq 10 \| xargs -Iz ./ttfb.sh "https://google.com"

	curl -o /dev/null \
	-H 'Cache-Control: no-cache' \
	-s \

	###################################################
	##
	## Alertmanager YAML configuration for routing.
	##
	## Will route alerts with a code_owner label to the slack-code-owners receiver
	## configured above, but will continue processing them to send to both a
	## central Slack channel (slack-monitoring) and PagerDuty receivers
	## (pd-warning and pd-critical)
	##