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.

chandika

Executive Summary

This guide documents how to use Factory's Droid CLI with your Claude Code Max subscription (OAuth authentication) instead of pay-per-token API keys. The solution leverages CLIProxyAPI as a transparent authentication proxy that converts API key requests from Factory CLI into OAuth-authenticated requests for Anthropic's API.

Architecture Overview

Factory CLI → [Anthropic Format + API Key] → CLIProxyAPI → [Anthropic Format + OAuth] → Anthropic API
                                                  ↓
 (Auth Header Swap)

irhum

# build your encoder upto here. It can simply be a series of dense layers, a convolutional network
# or even an LSTM decoder. Once made, flatten out the final layer of the encoder, call it hidden.

# we use Keras to build the graph

latent_size = 5
mean = Dense(latent_size)(hidden)

# we usually don't directly compute the stddev σ 
# but the log of the stddev instead, which is log(σ)

StuartGordonReid

import math
import numpy
import numpy.random as nrand

"""
Note - for some of the metrics the absolute value is returns. This is because if the risk (loss) is higher we want to
discount the expected excess return from the portfolio by a higher amount. Therefore risk should be positive.
"""