Vectra AI is a leader in the cybersecurity field, leveraging the capabilities of artificial intelligence (AI) to provide clients with cybersecurity threat detection and real-time response across all domains of enterprise systems. Backed by a decade of experience, Vectra AI protects over seven million hosts and 1700 customers.

A premier priority of Vectra AI is to identify cyber threats through analysis of network activity. Doing this effectively requires knowing what devices are connected to that network. Vectra AI currently relies on the Media Access Control (MAC) address, a unique number assigned to a computer chip, to distinguish devices from one another. However, this MAC can be faked, and without reliable identification, a customer may not recognize an unfamiliar device until it is too late.

To eliminate this gap, our AI-Powered Passive Asset Fingerprinting System uses various fingerprints derived from device characteristics left behind on the network. Our system combines characteristics that traditional rule-based systems struggle with.

Our system uses a hybrid of rule-based, machine learning, and large language models to extract new information from customer assets and make classifications. This approach benefits from both the performance of rule-based systems and the complexity of machine learning models. Over time, our system adds more verified device data and uses it to classify future unknown devices.

Our system utilizes an intuitive dashboard to enable Vectra AI users to visualize the network traffic. They are provided with at-a-glance statistics so they may make informed decisions about important security choices.

The AI-Powered Passive Asset Fingerprinting System uses a React with TypeScript front end. It accesses our Python and FastAPI back end through REST API, and stores data into our PostgreSQL database.