Midex AI — a visual interface for managing AI agents and data flows in the TON ecosystem

During the TON AI Hackathon, I built an infrastructure-level visual layer for managing AI agents on TON

Detailed submission for the Agent Infrastructure track:
https://identityhub.app/contests/ai-hackathon?submission=cmmr8nmet007q01qxzd0z4eep
Demo video: https://youtu.be/SzGCOAEAdas

Detailed submission for the User-Facing AI Agents track:
https://identityhub.app/contests/ai-hackathon?submission=cmn5ib9ou00cp01nvlv807mve
Demo video: https://www.youtube.com/watch?v=tfx7cqHY9dM

This is already a working system, ready to scale.

👇 YOU CAN:

• connect APIs, Telegram bots, and any external data sources
• create and train AI agents for specific tasks
• build complex flows from nodes and manage them directly from Telegram
• AI agents can analyze data, make decisions, and execute actions autonomously

USE CASE:

• trading
• analytics
• research
• process automation

Midex AI