Computer vision loss prevention for retail and convenience stores
Panoptyc deploys machine learning and computer vision to detect theft and suspicious behavior in grocery stores, convenience stores, and micro markets. The tech stack—NVIDIA Jetson, TensorRT, ONNX Runtime, AWS IoT Core, and Greengrass—reveals a heavy edge-computing architecture optimized for on-premises inference rather than cloud-dependent processing, critical for retail environments with inconsistent connectivity. Active projects around edge device platform design, AWS IoT integration, and model inference optimization confirm they're scaling hardware deployment and real-time detection at the edge. Sales-heavy hiring (26 of 50 open roles) signals aggressive go-to-market expansion.
Notable leadership hires: Lead Generation Specialist, Director of Sales
Panoptyc builds a loss prevention platform that uses machine learning and computer vision to identify theft and high-risk behavior across retail locations. The system operates on edge devices (Raspberry Pi, NVIDIA Jetson boards) deployed directly in stores, feeding into a real-time dashboard backend. The company targets grocery operators, convenience store chains, and micro market managers in the United States and Latin America. Infrastructure runs on AWS (IoT Core, Greengrass, containerization via Docker/Kubernetes), with a frontend built in Angular and React. The organization spans 51–200 employees across sales, engineering, and operations, with accelerating hiring velocity.
Edge inference (NVIDIA Jetson, TensorRT, ONNX Runtime), AWS IoT (Core, Greengrass), backend on Node.js and .NET, frontend in Angular and React, PostgreSQL for data, containerized via Docker and Kubernetes.
United States, Philippines, India, Mexico, and across Latin America (Argentina, Colombia, Chile, Brazil, Peru, Costa Rica) and Europe (Germany, France, Bulgaria, Belgium, Hungary, Lithuania, Romania).
Panoptyc's technology stack, projects, and hiring signals are inferred from public hiring and company data — career pages, public listings, and company web presence — then clustered and de-duplicated. Figures are estimates that refresh over time. Read our full methodology →
This is not an official vendor or customer list. It is a technology-adoption signal inferred from public data, intended for B2B research.