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.

mislav

# you must also upload the public key as signing key to https://github.com/settings/keys
git config --global user.signingkey '~/.ssh/id_rsa'
git config --global commit.gpgsign true
git config --global gpg.format ssh
# (optional) this file enables `git log --show-signature` and contains email addresses associated with public keys
git config --global gpg.ssh.allowedsignersfile '~/.ssh/git-signers'

hopsoft

Smart Heroku Review Apps managed by GitHub Actions

This gist aims to provide a simple solution for managing Heroku Review Apps with GitHub Actions due to the security incident that continues to disrupt Heroku's GitHub integration. Watch the demo to learn more.

.github
├── workflows
│   ├── heroku_review_app_create.yml

miguelmota

-- table is 'blocks'
-- column is 'number'

SELECT
       gap_start, gap_end FROM (
              SELECT number + 1 AS gap_start,
              next_nr - 1 AS gap_end
              FROM (
                     SELECT number, lead(number) OVER (ORDER BY number) AS next_nr
                     FROM blocks

anvk

--- PSQL queries which also duplicated from https://github.com/anvk/AwesomePSQLList/blob/master/README.md
--- some of them taken from https://www.slideshare.net/alexeylesovsky/deep-dive-into-postgresql-statistics-54594192

-- I'm not an expert in PSQL. Just a developer who is trying to accumulate useful stat queries which could potentially explain problems in your Postgres DB.

------------
-- Basics --
------------

-- Get indexes of tables

camsaul

;;; -*- tab-width: 2 -*-

;;; These four lines go at the beginning of almost every code file. 16-byte iNES header

  .inesprg 1                      ; one bank of program code
  .ineschr 1                      ; one bank of picture data
  .inesmap 0                      ; we use mapper 0
  .inesmir 1                      ; Mirror setting always 1

;;; BANKING

jfrost

-- Completely unused indexes:
SELECT relid::regclass as table, indexrelid::regclass as index
     , pg_size_pretty(pg_relation_size(indexrelid))
  FROM pg_stat_user_indexes
  JOIN pg_index
 USING (indexrelid)
 WHERE idx_scan = 0
   AND indisunique IS FALSE order by pg_relation_size(indexrelid);