Devansh Kandpal 0xDevansh

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.

Exporting Modules and Functions from Python `init.py`

Any directory with __init__.py is considered a package in Python.

Any python files inside a package is considered a module.

Modules contain functions and other bindings that is always exported.

If you are outside the package, and you want to import a module from a package:

	import json
	import sqlite3
	import traceback
	from contextlib import closing

	global init_count
	STEP_SIZE = 1000
	with open('count_a.txt', 'r') as file:
	init_count = 0
	count = file.read().rstrip()

Devansh Kandpal 0xDevansh

LLM Wiki

The core idea

Exporting Modules and Functions from Python __init__.py

Exporting Modules and Functions from Python `init.py`