Julio Camacho@Juliocamach

The Problem: In industrial environments, undetected pressure drops can lead to critical system failures and production downtime. Traditionally, these systems rely on basic PID controllers that lack contextual awareness. The Solution: This project introduces an AI Agent to replace traditional controllers, bridging the gap between advanced Large Language Models and physical hardware. Built for predictive maintenance, the system continuously monitors telemetry and makes real-time, autonomous decisions to stabilize industrial infrastructure. How it Works: An ESP32 microcontroller at the edge reads analog pressure data and streams telemetry via the MQTT protocol. A Python-based agent intercepts this data and queries the Qwen2.5-7B-Instruct model, deployed via vLLM on the AMD Developer Cloud. Acting strictly on mechatronic safety constraints, the AI evaluates the pressure. If it drops below the critical 15.0 PSI threshold, the agent overrides the system into "Alert" mode and commands the hardware to activate an auxiliary pump at 100% PWM. A Streamlit dashboard visualizes the live data and automatically generates an audit log. Impact: By utilizing the ultra-low latency of the AMD Cloud, this proof of concept demonstrates that LLMs can actively govern physical hardware in Industry 4.0.

Grab your ideas, spin up compute, and start building what’s next! Start with $100 in AMD Developer Cloud credits and begin building right away. Get access to hands-on tutorials, workshops, and curated learning resources to help you build faster and go further. To unlock credits and join the hackathon, sign up for the AMD AI Developer Program. 🤝 Go solo or team up — ship something real. 💰 $21,500+ total prize pool & 1 × AMD Radeon™ AI PRO R9700 GPU up for grabs. 📍 On-site Venue (May 9-10): MindsDB SF AI Collective 3154 17th St, San Francisco, California, USA 📍On-site participation is by invitation only. Travel and accommodation expenses will not be covered. 🧑‍💻 Developers, builders, and innovators - bring your boldest ideas and let’s turn them into real AI systems on AMD.