Marcos Hidalgo Flores@josuehidalgoflores58498

In Peru's electronic security sector, companies that install CCTV, access control, and fire detection systems at retail sites are only paid after submitting a formal compliance report with visual evidence for every device. Today, an assistant spends 12+ hours per site manually collecting photos from chaotic group chats, pairing before/after evidence, and writing observations — a critical bottleneck that delays cash flow. Centinela Agent integrates directly into the messaging apps field technicians already use on-site. As crews document their work by sending photos and captions, the agent ingests each image through a CPU preprocessing pipeline — compression, perceptual hashing, blur detection — clusters near-duplicates, and queues only the sharpest representative per group for GPU inference. A Gemma 4 31B multimodal model served via vLLM on an AMD Instinct MI300X (192 GB HBM3) classifies each photo: equipment category, intervention state (in-progress vs. completed), floor plan location, OCR data, and technician observations. Anomalies trigger a Human-in-the-Loop confirmation flow with the field supervisor. When a project closes, a structured compliance report is generated automatically. What took 12 hours now takes under 15 minutes of human review.

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.