AI-powered social engagement tool for solo builders and indie hackers.
"AI’s gentle touch, engaging mind — appreciating voices, promoting dreams."
ViralWhisper is an AI-powered assistant built for indie devs, solo founders, and product builders who are deeply involved in development but don’t have time to engage on platforms like Twitter/X and Reddit.
It automatically:
- Finds relevant posts in your niche
- Understands the context using local LLMs (Ollama, etc.)
- Generates meaningful replies
- Adds a subtle promotional line about your product or upcoming launch
With ViralWhisper, your presence grows — even when you're heads-down building 🚧
Building in public is powerful, but:
- It's time-consuming to engage daily on Reddit/X
- Context-aware replies are hard to automate
- Promoting your product can feel spammy
ViralWhisper solves all of this by:
- Using keywords and niches to track relevant content
- Parsing posts using AI (locally run LLMs like
DeepSeek-r1,mistral) - Replying with human-like, value-adding messages
- Appending soft promotional CTAs based on your config
All while you focus on shipping 🚢
-
Input: You define your:
- Keywords (e.g.,
AI,startup,web3) - Product description, promo lines, CTA
- Keywords (e.g.,
-
Fetch: Bot scrapes relevant posts from Reddit and Twitter (coming soon)
-
Process: Uses Ollama LLM locally to understand post context
-
Generate: Crafts a thoughtful reply + soft plug for your product
-
Engage: Posts response via Reddit API (Twitter support coming)
- ✅ Reddit integration with keyword-based filtering
- ✅ Local AI response generation (Ollama, no OpenAI keys needed)
- ✅ Auto-logging of interactions
- ✅ Configurable promotional CTA
- 🚧 Twitter/X support (Coming Soon)
- 🚧 Frontend for easier config (Planned)
Python, Reddit API (PRAW), Ollama (LLMs), CSV, ConfigParser, Logging
reddit-bot/
├── reddit_promo_bot.py # Main bot logic
├── config.ini # Config: keywords, promo text, etc.
├── reddit_interactions.csv # Logs for posts replied to
├── bot-1.py # Extra bot logic/testing
├── reddit_bot.log # Logging
└── README.md
-----