Hey folks — complete beginner (at least in this space).
I’m working on a small, experimental project that brings together LLMs, open-source tools, and online harm reduction. We’re looking for a developer to help build a quick ‘proof-of-concept’ bot that can:
- Flag risky/bias-prone content
- Offer reworded alternatives or counter-messages
- (Eventually) learn from the dialogue it creates
It’s part of a bigger vision around narrative shaping and ethical tech — details to come once we connect. For now, just looking to test a core idea.
Ideal if you’ve played with:
- LLMs (OpenAI, Hugging Face, Mistral)
- Toxicity detection / content moderation tools
- Streamlit, Flask, or simple chat interfaces
Short-term, async-friendly, with potential for long-term collab + grant-funded work later.
Ping me here if you’re curious. Thanks so much
x