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.

ctoth

Working with Q — Coding Agent Protocol

What This Is

Applied rationality for a coding agent. Defensive epistemology: minimize false beliefs, catch errors early, avoid compounding mistakes.

This is correct for code, where:

• Reality has hard edges (the compiler doesn't care about your intent)
• Mistakes compound (a wrong assumption propagates through everything built on it)
• The cost of being wrong exceeds the cost of being slow

renshuki

Elasticsearch Painless script which aims to calculate the difference in days between a date indexed into a document and the current date.

GET days_compare/_search
{
  "script_fields": {
    "diffdate": {
      "script": {
        "lang":   "painless",
 "source": """

Mike-Gough

React Native App Creation Recipe

This is a step-by-step guide to create React Native app.

You will get an application which has;

• TypeScript
• Linting
• Formatting
• Testing

ericmjl

UPDATE: I have baked the ideas in this file inside a Python CLI tool called pyds-cli. Please find it here: https://github.com/ericmjl/pyds-cli

How to organize your Python data science project

Having done a number of data projects over the years, and having seen a number of them up on GitHub, I've come to see that there's a wide range in terms of how "readable" a project is. I'd like to share some practices that I have come to adopt in my projects, which I hope will bring some organization to your projects.

Disclaimer: I'm hoping nobody takes this to be "the definitive guide" to organizing a data project; rather, I hope you, the reader, find useful tips that you can adapt to your own projects.

Disclaimer 2: What I’m writing below is primarily geared towards Python language users. Some ideas may be transferable to other languages; others may not be so. Please feel free to remix whatever you see here!

gvolpe

Dependency Injection in Functional Programming

There exist several DI frameworks / libraries in the Scala ecosystem. But the more functional code you write the more you'll realize there's no need to use any of them.

A few of the most claimed benefits are the following:

• Dependency Injection.
• Life cycle management.
• Dependency graph rewriting.

gricard

If you enjoyed reading this, I'm intending to do more blogging like this over here: https://cdgd.tech

This is not a complaint about Webpack or v4 in any way. This is just a record of my process trying it out so I could provide feedback to the webpack team

Hmm... I don't see any docs for 4.0 on https://webpack.js.org. I guess I'll just wing it. All I need to do is npm i -D webpack@next, right?

+ [email protected]

robertobarlocci

From -> https://forum.synology.com/enu/viewtopic.php?t=98124

1. Go to your console ( https://console.developers.google.com/ ) and create a project.
2. Create a bucket in Storage -> Cloud Storage -> Storage Browser
3. Go to Storage -> Cloud Storage -> Storage Access and enable Interoperability
4. Generate access key
5. Now go to DSM 5.2 and install Hyper Backup if you havent already.
6. Pick S3 storage
7. The input next to the label "S3 Server" is both text input and drop-down. This is where you write "storage.googleapis.com"
8. Enter key and secret.

ageis

YubiKey 4 series GPG and SSH setup guide

Written for fairly adept technical users, preferably of Debian GNU/Linux, not for absolute beginners.

You'll probably be working with a single smartcard, so you'll want only one primary key (1. Sign & Certify) and two associated subkeys (2. Encrypt, 3. Authenticate). I've published a Bash function which automates this slightly special key generation process.

clarkbw

const timing = store => next => action => {
  performance.mark(`${action.type}_start`);
  let result = next(action);
  performance.mark(`${action.type}_end`);
  performance.measure(
    `${action.type}`,
    `${action.type}_start`,
    `${action.type}_end`
  );
  return result;