karpathy

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.

Amerousful

Gatling: How to pass the session variables between two scenarios?

I have a general solution which can easy to use. There are several steps how to do it.

Create methods which get all session variable from one scenario and another method which sets this variables:

import scala.collection.concurrent.{Map => ConcurrentMap}

def passAnotherSession(implicit transferVariable: ConcurrentMap[String, Any]): Expression[Session] =

sergeyklay

Journalctl Cheat Sheet

Configuration

Permissions

To see messages from other users and the system as well as performing various log operations from a regular user add it to the group:

sudo usermod -aG systemd-journal $USER

sergeyklay

Happy PHP Debugging

Usage

gdb --args /usr/bin/php script.php

(gdb) source ~/src/php/7.2.1/.gdbinit

sergeyklay

Multiple PHP versions using phpenv and php-build

Install dependecies

Debian/Ubuntu users

sudo apt install \
  autoconf \
  bison \

sergeyklay

cmake_minimum_required(VERSION 3.5)

project(extname
          VERSION 1.0.0
          LANGUAGES C)

message(STATUS "Begin cmaking of PHP extension ...")

if (NOT CMAKE_BUILD_TYPE)
    set(CMAKE_BUILD_TYPE Debug CACHE STRING