Skip to content

Instantly share code, notes, and snippets.

View sumitsahoo's full-sized avatar
👨‍💻
Architecting…

Sumit Sahoo sumitsahoo

👨‍💻
Architecting…
View GitHub Profile
@sumitsahoo
sumitsahoo / add-to-github.sh
Created September 4, 2025 07:33
Add existing repository to GitHub
#!/bin/bash
# ==============================================================================
# A simple script to link a local Git repository to a new remote on GitHub
# and push the initial 'main' branch.
#
# INSTRUCTIONS:
# 1. Make sure you have initialized a Git repository (`git init`).
# 2. Add and commit your files (`git add .` and `git commit -m "Initial commit"`).
# 3. Replace the URL with your own repository's URL.
@sumitsahoo
sumitsahoo / git_push.sh
Created March 4, 2026 10:16
Push an existing repository to GitHub
# push an existing repository from the command line
git remote add origin https://github.com/username/reponame
git branch -M main
git push -u origin main
@sumitsahoo
sumitsahoo / python_setup.md
Created April 13, 2026 15:18
Setup Python on Mac using UV

🐍 Install Python on macOS using uv

📋 Prerequisites

  • 🍎 macOS with Homebrew installed
  • 💻 Terminal (zsh)

🚀 Step 1: Install uv

@sumitsahoo
sumitsahoo / llm-wiki.md
Created April 13, 2026 16:02 — forked from karpathy/llm-wiki.md
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.