Andreas Sefzig sefzig

LLM Wiki v2

A pattern for building personal knowledge bases using LLMs. Extended with lessons from building agentmemory, a persistent memory engine for AI coding agents.

This builds on Andrej Karpathy's original LLM Wiki idea file. Everything in the original still applies. This document adds what we learned running the pattern in production: what breaks at scale, what's missing, and what separates a wiki that stays useful from one that rots.

What the original gets right

The core insight is correct: stop re-deriving, start compiling. RAG retrieves and forgets. A wiki accumulates and compounds. The three-layer architecture (raw sources, wiki, schema) works. The operations (ingest, query, lint) cover the basics. If you haven't read the original, start there.

This is an OPML version of the HN Popularity Contest results for 2025, for importing into RSS feed readers.

Plug: if you want to find content related to your interests from thousands of obscure blogs and noisy sources like HN Newest, check out Scour. It's a free, personalized content feed I work on where you define your interests in your own words and it ranks content based on how closely related it is to those topics.

<core_identity> You are an assistant called Cluely, developed and created by Cluely, whose sole purpose is to analyze and solve problems asked by the user or shown on the screen. Your responses must be specific, accurate, and actionable. </core_identity>

<general_guidelines>

NEVER use meta-phrases (e.g., "let me help you", "I can see that").
NEVER summarize unless explicitly requested.
NEVER provide unsolicited advice.
NEVER refer to "screenshot" or "image" - refer to it as "the screen" if needed.
ALWAYS be specific, detailed, and accurate.

Anti-hype LLM reading list

Goals: Add links that are reasonable and good explanations of how stuff works. No hype and no vendor content if possible. Practical first-hand accounts of models in prod eagerly sought.

Foundational Concepts

Pre-Transformer Models

I combined Mike Bostock's Zoomable Sunburst and Sunburst Partition examples, so that I could have both zooming and updating the underlying data (between count and size, in this case). A simple combination of the examples does not work; you have to edit the arcTween function used for updating the data, so that when it redraws the partition layout, it takes account of the current zoom level by adjusting the domain of the x scale.

Click on any arc to zoom in, and click on the center circle to zoom out. Use the Size/Count radio buttons to update the data.

	"""
	The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
	This file is the complete algorithm.
	Everything else is just efficiency.

	@karpathy
	"""

	import os # os.path.exists
	import math # math.log, math.exp

	// Inspired by https://twitter.com/coderitual/status/1112297299307384833 and https://tapajyoti-bose.medium.com/7-killer-one-liners-in-javascript-33db6798f5bf

	// Remove any duplicates from an array of primitives.
	const unique = [...new Set(arr)]

	// Sleep in async functions. Use: await sleep(2000).
	const sleep = (ms) => (new Promise(resolve => setTimeout(resolve, ms)));
	// or
	const sleep = util.promisify(setTimeout);

	var fileRelativeUrl = "Shared%20Documents/book1.xlsx";
	var sheetName = "Sheet1";
	var range = "A1\|J45";

	var url = _spPageContextInfo.webServerRelativeUrl.replace(/\/$/, '') + "/";
	loadXMLDoc(url + "_vti_bin/ExcelRest.aspx/" + fileRelativeUrl + "/model/Ranges('" + sheetName + "!" + range + "')?$format=atom", function (text) {
	var parser = new DOMParser();
	var excelRestNS = 'http://schemas.microsoft.com/office/2008/07/excelservices/rest';
	var xmlDoc = parser.parseFromString(text, "text/xml");
	var rows = xmlDoc.getElementsByTagNameNS(excelRestNS, 'row');

	/*

	This script is meant to be used with a Google Sheets spreadsheet. When you edit a cell containing a
	valid CSS hexadecimal color code (like #000 or #000000), the background color will change to that
	color and the font color will be changed to the inverse color for readability.

	To use this script in a Google Sheets spreadsheet:
	1. go to Tools » Script Editor;
	2. replace everyting in the text editor with this code;
	3. click File » Save;

	{
	"name": "Root",
	"children": [
	{
	"name": "Branch 1",
	"children": [
	{"name": "Leaf 3"},
	{"name": "Leaf 4"}
	]
	},