Skip to content

Instantly share code, notes, and snippets.

View iki's full-sized avatar

Jan Killian iki

View GitHub Profile
@rohitg00
rohitg00 / llm-wiki.md
Last active May 5, 2026 02:36 — forked from karpathy/llm-wiki.md
LLM Wiki v2 — extending Karpathy's LLM Wiki pattern with lessons from building agentmemory

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.

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.

@jvgrootveld
jvgrootveld / nextjs-firebase-admin-sdk-api-auth-middleware.ts
Last active June 24, 2024 10:25
API Auth middleware for NextJS with firebase admin SDK as this does not work with _middleware files
import { auth } from '../../firebase/firebase-admin'
import { NextApiRequest, NextApiResponse } from 'next'
type NextContextApiHandler = (req: NextApiRequestWithContext, res: NextApiResponse) => Promise<void>
export interface Context {
uid: string
}
export interface NextApiRequestWithContext extends NextApiRequest {
@TarlogicSecurity
TarlogicSecurity / kerberos_attacks_cheatsheet.md
Created May 14, 2019 13:33
A cheatsheet with commands that can be used to perform kerberos attacks

Kerberos cheatsheet

Bruteforcing

With kerbrute.py:

python kerbrute.py -domain <domain_name> -users <users_file> -passwords <passwords_file> -outputfile <output_file>

With Rubeus version with brute module:

@tamlyn
tamlyn / README.md
Last active July 7, 2022 09:48
Execution order of Jest/Jasmine test code

Execution order of Jest/Jasmine test code

While tests run in source order, surrounding code does not which can lead to hard to debug issues.

Compare the test file below with the sample output below that and note the order of the log messages.

Key points

  • Any code not inside of it, beforeAll, afterAll, beforeEach or afterEach runs immediately on initialisation.
  • This means code at the end of your file runs before even your before hooks.

Reviewer's checklist - How to boost your code review skills

Purpose of my lightning talk is to present insights I've reached as a reviewer/ reviewee in the last couple of months.

I will answer these questions:

  • What makes a good code review good?
  • How to build guidelines? Dos and Don'ts.
  • How to educate dev/team members about the code review process?
@tigerhawkvok
tigerhawkvok / restartAudio.bat
Created March 4, 2016 20:30
Restart Windows Audio Service (with admin permissions, if needed)
rem Written for buggy audio drivers that need to be restarted
rem In my case, the Claro 8.18 drivers bug out every once in a while on Windows 10, and need restarting to not sound poppy
@echo off
goto check_Permissions
:check_Permissions
echo Administrative permissions required to run this script. Checking...
net session >nul 2>&1
@azu
azu / Medium: remove location hash.user.js
Created December 2, 2015 15:05
Medium: remove location hash
// ==UserScript==
// @name Medium: remove location hash
// @namespace http://efcl.info/
// @description Remove location hash from medium
// @include https://medium.com/*#*
// @version 1
// @grant none
// ==/UserScript==
function removeLocationHash(){
@mistic100
mistic100 / Mediakeys.ahk
Last active March 22, 2026 04:32
Media keys shortcuts for AutoHotkey
; AutoHotkey Media Keys
^!Space::Send {Media_Play_Pause}
^!Left::Send {Media_Prev}
^!Right::Send {Media_Next}
^!NumpadMult::Send {Volume_Mute}
^!NumpadAdd::Send {Volume_Up}
^!NumpadSub::Send {Volume_Down}
@paulirish
paulirish / what-forces-layout.md
Last active May 4, 2026 19:42
What forces layout/reflow. The comprehensive list.

What forces layout / reflow

All of the below properties or methods, when requested/called in JavaScript, will trigger the browser to synchronously calculate the style and layout*. This is also called reflow or layout thrashing, and is common performance bottleneck.

Generally, all APIs that synchronously provide layout metrics will trigger forced reflow / layout. Read on for additional cases and details.

Element APIs

Getting box metrics
  • elem.offsetLeft, elem.offsetTop, elem.offsetWidth, elem.offsetHeight, elem.offsetParent