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.

alexlitz

#!/usr/bin/env python3
"""
TinyAdder: 36-parameter hand-crafted transformer for 10-digit addition.

Parameter counting:
- Identity mappings (direct copy): 0 params
- Broadcast (1 value to N outputs): 1 param
- Distinct values: count each
"""
import torch

cosminscn

"""
 Dynamic NanoGPT Adder: 130 params + 0 buffers

    transformer.wte.A                           [10, 1] =  10
    transformer.wte.B                            [1, 4] =   4
    transformer.h.0.attn.c_attn.weight          [12, 4] =  48
    transformer.h.0.attn.c_proj.weight           [4, 4] =  16
    transformer.h.0.mlp.c_fc.weight              [4, 4] =  16
    transformer.h.0.mlp.c_fc.bias                   [4] =   4
    transformer.h.0.mlp.c_proj.u                 [4, 1] =   4

karpathy

"""
The most atomic way to train and run inference for a GPT in pure, dependency-free Python.
This file is the complete algorithm.
Everything else is just efficiency.

@karpathy
"""

import os       # os.path.exists
import math     # math.log, math.exp

neudinger

#include <ranges>

#include <print>

#include <cctype>
#include <string>
#include <string_view>

using namespace std::literals;
using namespace std::string_view_literals;

lodestone-rock

import os
import wandb
from tqdm import tqdm
import argparse

import torch
import torchvision
import torch.nn as nn
import torch.nn.functional as F
from torch.utils.data import DataLoader

rekram1-node

{
  "agent_servers": {
    "OpenCode": {
      "command": "opencode",
      "args": ["acp"]
    }
  }
}

glennmatlin

Claude Code Hooks for working with uv

by Glenn Matlin / glennmatlin on all socials

What This Does

Prevents Claude Code from using pip, python, pytest, etc. directly in projects that use uv for Python environment management. Instead, Claude is guided to use uv run, uv add, etc.

Smart detection:

• Only activates in directories with pyproject.toml

yanndebray

philschmid

import os
from google import genai
from google.genai import types

client = genai.Client(api_key=os.getenv("GEMINI_API_KEY","xxx"))

# Repalce with the youtube url you want to analyze
youtube_url = "https://www.youtube.com/watch?v=RDOMKIw1aF4" 

# Prompt to analyze and summarize the Youtube Video