Skip to content

Instantly share code, notes, and snippets.

View ankushKun's full-sized avatar
💡
just keep building

Ankush Singh ankushKun

💡
just keep building
273 followers · 282 following
View GitHub Profile
@karpathy
karpathy / llm-wiki.md
Created April 4, 2026 16:25
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.

@ankushKun
ankushKun / canvas-renderer.html
Created December 24, 2025 14:17
Turn a 2D canvas into a 3D scene using clever maths
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Engine</title>
<style>
body {
margin: 0;
@ankushKun
ankushKun / ao-bot-deathmatch-betteridea.md
Last active April 5, 2025 04:59
@ankushKun
ankushKun / blockchain.cpp
Last active February 1, 2024 08:54
Smol project we were forced to make during class
////////////////////////////////////
// WORTHLESS BLOCKCHAIN PROJECT //
////////////////////////////////////
// The most basic imitation/implementation of how a Blockchain would work
// Still has lots of flaws (╯°□°）╯︵ ┻━┻
#include <ctime>
#include <cstring>
#include <iostream>
@ankushKun
ankushKun / SimpDetector.py
Created April 12, 2023 15:14
Simple Simp Detector 🫠
# DEPENDENCIES:
# pip3 install opencv-python deepface
# TODO:
# Add multithreading support so face detection is done in a different thread independent form video feed and display
# Add good support for multiple face detection (that is not linear)
# ヾ(=`ω´=)ノ” ..nyaa
import cv2
@ankushKun
ankushKun / SignTransactionExample.cs
Last active May 18, 2022 07:31
Sign DeSo transaction hex in C# using Bouncy Castle cryptography library
using System;
using System.Linq;
using System.Text;
using Desonity;
using Org.BouncyCastle.Asn1;
using Org.BouncyCastle.Asn1.Sec;
using Org.BouncyCastle.Crypto.Digests;
using Org.BouncyCastle.Crypto.Parameters;
using Org.BouncyCastle.Crypto.Signers;
using Org.BouncyCastle.Math;
@kanshi
kanshi / validate-deso-identity-jwt.sc
Created March 30, 2022 00:13
import pdi.jwt.{Jwt, JwtAlgorithm}
import org.bitcoins.core.util.Base58
import org.bouncycastle.jce.spec.ECPublicKeySpec
import org.bouncycastle.jce.ECNamedCurveTable
import org.bouncycastle.jce.provider.BouncyCastleProvider
import java.security.{KeyFactory, PublicKey, Security}
import scala.util._
val jwtToken = "eyJhbGciOiJFUzI1NiIsInR5cCI6IkpXVCJ9.eyJpYXQiOjE2NDg1OTcwNzIsImV4cCI6MTY0ODU5NzY3Mn0.MFEtqJzrLTnjYwH4gAOsH-0qsrQIFsKcYHyHdE1WdWHLn_xa-fGPoV3OkDLr0Rk1IcIoawSrDXNxkyZR0wGhFw"
val desoPKey = "BC1YLhwpmWkgk2iM9yTSxzgUVhYjgessSPTiVHkkK9pMrhweqJnWrvK"
@sungkhum
sungkhum / jwt.py
Last active May 24, 2022 04:26
Generate JWT from SEEDHEX for BitClout DeSo in Python
import jwt
import binascii
from ecdsa import SigningKey, SECP256k1
''' SEEDHEX should always be kept private. It has access to your complete wallet. It's kinda like
seed phrase. This is why writing methods in backend isn't a good practice until we have derived keys.
You can only automate your own account and can't have user authorisation. It is recommended to use test account while using write methods.
You can find the seedHex of your account in your browser storage. Just open https://bitclout.com/ > Dev tools > Application > Storage > Local Storage > https://identity.bitclout.com > users > Select the public key with which you want to post > seedHex'''
@alschmiedt
alschmiedt / gist:2085af32381db89d1f1ca80d5883c0d5
Created June 2, 2021 15:43
Blockly Toolbox in JSON
{
"contents": [
{
"kind": "CATEGORY",
"contents": [
{
"kind": "BLOCK",
"blockxml": {},
"type": "controls_if"
},
@emilianavt
emilianavt / BestVTuberSoftware.md
Last active June 16, 2026 09:41
Best VTuber Software

Best VTuber software

This is a list of the most commonly used and relevant vtubing software. The "best" will always be subjective and depend on your specific requirements. Overall, the information in this list is as accurate as I could figure it out, but there might be errors or some details might become out of date. If you find anything that needs to be corrected, please let me know. You can also note it in a comment.

Additional explanations:

  • iPhone means that an iPhone is basically required
  • iFacialMocap support means that tracking data can be received from the iFacialMocap iPhone app
  • VMC protocol means that the application can send and/or receive tracking data from other VMC protocol capable applications, allowing the combination of multiple tracking methods (e.g. VSeeFace receiving VR tracking from Virtual Motion Capture and iPhone/ARKit face tracking from Waidayo)
  • Tobii means that the Tobii eye tracker is supported