Skip to content

Instantly share code, notes, and snippets.

View gmcabrita's full-sized avatar

Gonçalo Mendes Cabrita gmcabrita

View GitHub Profile
@ry
ry / llm-wiki.md
Last active November 2, 2026 18:08 — forked from karpathy/llm-wiki.md
llm-wiki

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.

@mattpocock
mattpocock / README.md
Created July 7, 2026 11:51
agent-proxy — a zero-dep proxy that shows the bloat in Claude Code's requests (ranked tool table + full readable Markdown of every request)

agent-proxy — see the bloat in Claude Code's requests

A zero-dependency logging proxy that sits between Claude Code and the Anthropic API. It forwards every request untouched, streams the reply straight back (so the CLI is unaffected), and writes a readable Markdown document for each request — led by a ranked table of what is eating your context.

Run it

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.

defmodule Task do
defp await_one(tasks, timeout \\ 5_000) when is_list(tasks) do
awaiting =
Map.new(tasks, fn %Task{ref: ref, owner: owner} = task ->
if owner != self() do
raise ArgumentError, invalid_owner_error(task)
end
{ref, true}
end)
package examples
import (
"github.com/XANi/loremipsum"
"github.com/go-chi/chi/v5"
. "github.com/starfederation/datastar-dev/site/shared"
datastar "github.com/starfederation/datastar/sdk/go"
"math/rand/v2"
"net/http"
"time"
@gurgeous
gurgeous / justfile
Created December 7, 2024 01:55
Sample rails+vite justfile (see https://github.com/casey/just)
# add node_modules to $PATH
export PATH := justfile_directory() + "/node_modules/.bin:" + env_var('PATH')
# first recipe is always default - list these rules
default:
@just --list
# everything ok? good to run before commit
check:
@for i in tsc lint test vitest; do \
@stenuto
stenuto / hls.sh
Created November 7, 2024 16:58
HLS ffmpeg script
#!/bin/bash
# Function to display usage information
usage() {
echo "Usage: $0 /path/to/input.mp4 [ /path/to/output_directory ]"
exit 1
}
# Check if at least one argument (input file) is provided
if [ $# -lt 1 ]; then
@mmozeiko
mmozeiko / _miniperf_readme.md
Last active July 19, 2026 14:56
get PMU counter values with ETW, perf or kperf

MiniPerf

Example of how to capture CPU counters with ETW on Windows, perf on Linux or kperf on Apple.

Using ETW needs somewhat recently updated Windows 10 or 11. Not sure about exact version.

Currently tested on:

  • etw on Qualcomm Snapdragon X Elite, Windows 11, arm64
  • etw on AMD Zen 3, Windows 11 (with virtualization enabled in BIOS)

Multiple GitHub accounts (Work vs Personal)

This setup uses some tricks to ensure that the right email/name/ssh-key is used for the right repos without having to think about it ever again.

  • First generate two SSH keys, ~/.ssh/id_ed25519 and ~/.ssh/id_ed25519_work
  • Add one key to your personal account and the other to your work account

.ssh/config

@mitchellh
mitchellh / merge_vs_rebase_vs_squash.md
Last active October 31, 2026 18:54
Merge vs. Rebase vs. Squash

I get asked pretty regularly what my opinion is on merge commits vs rebasing vs squashing. I've typed up this response so many times that I've decided to just put it in a gist so I can reference it whenever it comes up again.

I use merge, squash, rebase all situationally. I believe they all have their merits but their usage depends on the context. I think anyone who says any particular strategy is the right answer 100% of the time is wrong, but I think there is considerable acceptable leeway in when you use each. What follows is my personal and professional opinion: