Artavazd Balaian REASY

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.

Agent Guidelines for Rust Code Quality

This document provides guidelines for maintaining high-quality Rust code. These rules MUST be followed by all AI coding agents and contributors.

Your Core Principles

All code you write MUST be fully optimized.

"Fully optimized" includes:

AGENTS.md

Persona

Address the user as Cam.
Optimize for correctness and long-term leverage, not agreement.
Be direct, critical, and constructive — say when an idea is suboptimal and propose better options.
When writing work summeries or replying to user questions be sure to explain things in a clear and easy to understand language, don't be over-technical unless user asks for it, give code examples to help explain what you're refering to and provide context.

Quality

Inspect project config (package.json, etc.) for available scripts.

The Death of the User Interface

TL;DR: We're witnessing the end of graphical user interfaces. AI agents like Claude Code are eliminating the need for windows, menus, and clicks, replacing them with natural language. The computer is finally learning to speak human, not the other way around.

🔮 A Personal Revelation

Last week, I realized something profound: I haven't opened Finder in months. Not once.

Deploy Istio

VM setup

Start from a minimal installation of Ubuntu Server 22.04.

In this case, I've set up a VM with two network interfaces, one NAT for connection to the internet, and one bridged network interface to connect from the host to the guest VM.

MSI hash table benchmarks

These are benchmarks demonstrating claims made in [the quick and practical "MSI" hash table][blog]. It pits an MSI hash table set in C using an integer permutation hash and an AES-NI hash, a C++ std::set, and Go map[string]struct{}. Inputs are strings of length 1–8.

My results on an i7-8650U, GCC 12.0.1 (libstdc++), Clang 14.0.6 (libc++):

Time (s) Memory (MiB)

Patch APK

Full tutorial to patch APKs on Android using apk-mitm and APKLab. Support for bundle APKs.

Pull APK

adb shell pm list packages
adb shell pm path <package_name>

	[package]
	name = "testcontainers-sample"
	version = "0.1.0"
	edition = "2021"

	[dev-dependencies]
	async_once = "0.2.6"
	aws-sdk-s3 = "0.28.0"
	ctor = "0.2.4"
	lazy_static = "1.4.0"

	#![deny(clippy::all)]
	#![deny(clippy::nursery)]
	#![allow(clippy::cast_possible_truncation)]

	use std::{
	collections::hash_map::RandomState,
	hash::{BuildHasher, Hash, Hasher},
	};

	#[derive(Debug, Clone)]

	#!/bin/bash

	clean_debian() {
	[ "$MNT_DIR" != "" ] && chroot $MNT_DIR umount /proc/ /sys/ /dev/
	sleep 1s
	[ "$MNT_DIR" != "" ] && umount -l $MNT_DIR
	sleep 1s
	[ "$DISK" != "" ] && qemu-nbd -d $DISK
	sleep 1s
	[ "$MNT_DIR" != "" ] && rm -r $MNT_DIR