Fernando Paladini paladini

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.

Bank-Attested Age Verification (BAV)

A privacy-preserving, minimal-disclosure age check that reuses banks' existing KYC, with the user as the transport layer.

TL;DR

A bank authenticates an existing customer and signs a short-lived age-threshold claim - not their name, date of birth, account number, or other identity details.
The merchant binds that claim to its own nonce and a fresh WebAuthn credential, then atomically consumes the claim in a short-lived, single-use replay ledger. The reference verifier uses Redis.

🚀 Guia Definitivo para a POSCOMP: Todas as Provas e Gabaritos em um só lugar!

Você está se preparando para o Exame Nacional para Ingresso na Pós-Graduação em Computação (POSCOMP) e se sente perdido com tantos materiais espalhados pela internet?

Seus problemas acabaram! O repositório Awesome POSCOMP é um projeto de código aberto que reúne tudo o que você precisa para ser aprovado, de forma organizada e totalmente gratuita.

Chega de procurar por provas antigas em links quebrados. Foque no que realmente importa: os seus estudos!

Início Rápido: Removedor de Vocal e Separador de Instrumentos com IA

Uma ferramenta gratuita e de código aberto que transforma seu computador em um estúdio de separação de áudio com IA. Ela roda inteiramente na sua máquina local, o que significa sem APIs externas, sem assinaturas e sem custos. Separe vocais, bateria, baixo e outros instrumentos de qualquer arquivo de música ou link do YouTube usando o poderoso modelo de IA Demucs da Meta.

Documentação completa e código-fonte: https://github.com/paladini/voice-separator-demucs

🚀 Instalação Mais Fácil (Docker)

Quick Start: AI Vocal Remover & Instrument Separator

A free and open-source tool that turns your computer into an AI-powered audio separation studio. It runs entirely on your local machine, meaning no external APIs, no subscriptions, and no costs. Separate vocals, drums, bass, and other instruments from any song file or YouTube link using Meta's powerful Demucs AI model.

Full documentation and source code: https://github.com/paladini/voice-separator-demucs

🚀 Easiest Installation (Docker)

Semantic Commit Messages

See how a minor change to your commit message style can make you a better programmer.

Format: <type>(<scope>): <subject>

<scope> is optional

	#!/usr/bin/env python3
	"""
	Get Garmin OAuth tokens via real browser login (Playwright).
	Bypasses the 429-blocked SSO programmatic login endpoint.

	Usage:
	uv run --with playwright --with requests-oauthlib scripts/garmin-browser-auth.py

	First time setup (installs Chromium):
	uv run --with playwright python -m playwright install chromium

	You are ChatGPT, a large language model based on the GPT-5 model and trained by OpenAI.
	Knowledge cutoff: 2024-06
	Current date: 2025-08-08

	Image input capabilities: Enabled
	Personality: v2
	Do not reproduce song lyrics or any other copyrighted material, even if asked.
	You're an insightful, encouraging assistant who combines meticulous clarity with genuine enthusiasm and gentle humor.
	Supportive thoroughness: Patiently explain complex topics clearly and comprehensively.
	Lighthearted interactions: Maintain friendly tone with subtle humor and warmth.

	import csv
	import gzip
	import io
	import json
	from urllib.parse import urlencode, urljoin
	from urllib.request import Request, urlopen


	class BrasilIO:

	#Login as root
	sudo su

	#create jdk directory
	mkdir /opt/jdk

	#uncompress, change to your file name
	tar -zxf jdk-8u5-linux-x64.tar.gz -C /opt/jdk

	#check if files are there