Hey everyone!
I’ve been working on a security layer for the Agentic Economy during a hackathon, and I just hit a major milestone.
The problem: As AI agents start handling real money, they are becoming prime targets for "drainers" and sophisticated splitting attacks that traditional rule-based security misses.
The solution: ArcWarden & Imina Na. I’ve developed a vision-language security oracle. Instead of just looking at raw data, it "sees" transaction patterns.
The Tech Stack:
- Model: Fine-tuned Qwen2-VL (Vision-Language Model).
- Hardware: Trained on the beast AMD MI300X (ROCm).
- Dataset: 10,000+ transaction graph patterns (Dogon Dataset).
- Platform: Live dashboard (Sigui) connected to the Arc Testnet.
I just pushed the trained LoRA weights to Hugging Face! ????
I need your feedback! I’m looking for testers and devs to check out the dashboard and tell me what you think about using Vision AI for blockchain security. Can an AI "Oracle" actually stop the next big drainer?
???? Check the model on Hugging Face: https://huggingface.co/Ibonon/imina_na_lora
[link] [comments]
You can get bonuses upto $100 FREE BONUS when you:
💰 Install these recommended apps:
💲 SocialGood - 100% Crypto Back on Everyday Shopping
💲 xPortal - The DeFi For The Next Billion
💲 CryptoTab Browser - Lightweight, fast, and ready to mine!
💰 Register on these recommended exchanges:
🟡 Binance🟡 Bitfinex🟡 Bitmart🟡 Bittrex🟡 Bitget
🟡 CoinEx🟡 Crypto.com🟡 Gate.io🟡 Huobi🟡 Kucoin.
Comments