Slow & Steady: Building a Local AI Group Chat & Task Summarizer 🦥

For Hacktoberfest Week 3, I built an automation that embraces the "sloth mindset"—letting AI handle the tedious work of reading long messages and organizing tasks so you don't have to!

What I Built

A lightweight, 100% private local AI web app using Streamlit, Ollama, and Gemma. Instead of waking up to hundreds of unread WhatsApp or Telegram messages, this tool generates a quick 2-sentence morning summary along with an actionable to-do list for the day.

Key Features:

• Morning Briefing: Automatically summarizes overnight messages into a concise 2-sentence snapshot.
• Action Extraction: Identifies pending tasks, deadlines, and direct mentions so nothing gets missed.
• Offline & Private: Operates directly on the local machine using open-weight models via Ollama. Personal conversations never leave your device!

Tech Stack

• AI Engine: Gemma (run locally via Ollama)
• Frontend / Dashboard: Streamlit
• Language: Python

Code Repository

📁 GitHub Repo: https://github.com/gmcode-lab/slow-and-steady-.git

Why Open-Weight Models Matter

1. Absolute Privacy: Group chat logs contain private banter and contact details. Running Gemma locally ensures private messages stay strictly on your laptop.
2. Zero Operating Cost: No paid cloud APIs or token costs. Build once, run forever.
3. Offline Reliability: Works seamlessly without an active internet connection.

hacktoberfest #devchallenge #opensource #ai