We're building the UK’s sovereign visual autonomy layer — one edge-to-cloud platform stack that enables defence and security systems to think, adapt, and collaborate in real-time.
Ecosystem
ecosystem
FortifAI is our TRL 9 visual autonomy platform, delivering real-time threat analysis, sensor fusion and autonomous control. Scalable and versatile, it supports low-SWaP devices and high-performance AI servers and integrates seamlessly with third-party camera infrastructure and ML models. Global proven deployment across uncrewed land, air and sea platforms, as well as long-range surveillance systems.
UnifAI is a turnkey MLOps platform that provides real-time oversight and full lifecycle management for distributed AI systems at scale. It simplifies dataset management, device monitoring, and model deployment, while closing the loop between edge and cloud to enable scalable control, continuous improvement, and faster response across fleets of AI-enabled assets.
VizEdge is a rugged, wearable AI platform built for frontline use in communication-denied environments. Designed and manufactured in the UK, it delivers high-performance edge AI without bulky servers or connectivity. Its open architecture supports third-party models, applications and agents, giving teams the flexibility to adapt as mission needs evolve.
FortifAI runs on NVIDIA processors and connects seamlessly with defence and security cameras — from CCTV to uncrewed systems — through our growing set of industry-standard protocols for plug-and-play deployment.
Raw data into action, anywhere
Visible and thermal video feeds are processed by our modular AI stack, with resulting visual insights fused alongside radar and RF tracks — all without reliance on internet or GNSS.
Real-time AI asset coordination
Send threat alerts and metadata-tagged video to your C2 or VMS over low-bandwidth links. Each node curates its own training data, enabling continuous learning across the fleet.
Use cases
AI in Action: Proven Deployments
Unmanned Aerial Vehicle (UAV) Automation
FortifAI reduces pilot workload and boosts threat detection through a tightly integrated suite of visual AI capabilities. Whether deployed onboard an unmanned system, embedded in a drone dock, or linked to a ground controller, it delivers real-time geolocation of potential threats over land or sea, without GNSS, and can trigger automatic evasion when an interceptor is detected.
The Vizgard and Parrot combination has been very well received by our end users. Collaborating with Vizgard allowed us to consistently meet our tight project deadlines.
Richard, Principal Systems Engineer – DSTL
Long-range Maritime Security
FortifAI provides automated control of steerable long-range visible and thermal cameras to support surveillance at border, coastal and critical infrastructure sites. Tasks typically managed by centralised command and control systems, such as intelligent guard tours, automatic investigation and slewing to radar or RF tracks, can be offloaded to edge-AI nodes for faster, distributed and comms-resilient response. FortifAI also improves effective range through electronic stabilisation and real-time enhancement in fog, mist, smoke and other low-contrast environments.
“Vizgard impressed us during the first live demo at our test site, and we are in the early stages of working collaboratively on some of our projects.”
John, Engineering Director – Ultra
Uncrewed Surface Vessel (USV) Automation
FortifAI enables advanced autonomy on USVs by analysing multiple EO/IR camera feeds in real time to detect and track maritime objects such as speedboats, warships and buoys. Using NMEA-standard position data, it calculates true bearings and displays an augmented view with relative distance and AIS call signs. The system can autonomously follow or evade targets based on predefined rules (e.g. proximity or threat behaviour), or accept remote commands via satellite comms. All processing is done at the edge, reducing latency, conserving bandwidth and ensuring reliable operation in communication-denied environments.
We evaluated several low-latency computer vision solutions and chose Vizgard for our UAV and USV platforms. Their software, development approach, and mission-first mindset outperformed both SMEs and larger firms. Under pressure, Alex and the Vizgard team consistently deliver and lead in this field.
Ross, CEO & Founder – Volander
Airspace situational awareness
When integrated into a multi-sensor system such as a counter-drone solution or an unmanned traffic management platform, FortifAI can automatically steer multiple distributed cameras to visually confirm and differentiate between birds, drones, planes and helicopters. It estimates their approximate position and trajectory in 3D space, enhancing situational awareness and supporting real-time decision-making.
Vizgard were an integral part of the Skyway consortium. They are building pioneering technology important to establishing a unique, comprehensive surveillance airspace picture. We look forward to continuing to work together to enhance the technology and develop and strengthen our commercial relationship.
Richard, CEO & Founder – Altitude Angel
Perimeter Security
FortifAI provides a distributed edge-AI solution for automated control of steerable, long-range visible and thermal cameras. It automatically detects and tracks threats such as armed individuals or suspicious vehicles, while managing tasks that would normally require centralised command-and-control systems, including intelligent guard tours and automatic slew-to-cue investigations of radar or RF tracks. Multi-PTZ collaboration ensures faster responses and continuous, accurate visibility across borders, coastlines, and critical infrastructure sites.
When Vizgard first approached us in 2021, we were skeptical of AI after underwhelming results from other providers. However, Vizgard's AI exceeded our expectations. In live trials, our long-range EO/IR cameras with Vizgard AI classified people and vehicles at nearly 2 kilometers and even detected armed individuals. Vizgard is now a trusted partner, and we have recently used their AI to safeguard the perimeter of a major international airport.
Jubal, COO – Infiniti Electro-Optics
Drone ISR
The VizEdge is plug-and-play with FortifAI and Parrot Anafi USA or UKR quadcopters, allowing police or military teams to take high-performance AI into the field whether from Vizgard or a third party.
We’ve been burnt by AI providers who overpromise but underdeliver. Vizgard came highly recommended by a government partner, and they’ve delivered from day one — helping us win a bid and backing it up with proactive support. VizEdge performed well in trials and meets a clear need for a UK-made, rugged AI system that enhances automation and situational awareness on the go.
Sam, Global Business Dev Director – Coptrz
About
Our Story
Vizgard was born from Alex Kehoe’s time serving as a Royal Navy submariner, where he operated advanced weapon and navigation systems during covert missions and saw firsthand how sensor overload could lead to critical threats being missed. He later joined the innovation division of leading EU defence tech giant Hensoldt, where he helped pioneer early counter-drone and semi-automated surveillance solutions. Through global engagement with frontline users, he found that time-critical threat detection was, and still is, too labour-intensive. This challenge has only been made worse by the growing complexity of asymmetric threats.
Vizgard bridges the gap between the promise of AI-driven camera automation and the reality that widely used open-source models often underperform in real-world conditions. The future of the UK’s defence, security, and resilience depends on trusted, autonomy-enabled technology that can scale across domains and camera types. It must be delivered at pace and be software-defined to future-proof both legacy systems and those needed for emerging conflicts. Vizgard is dedicated to making that future a reality.
Vizgard Secures £1.5M in Over-subscribed Seed round
Events
March 14, 2025
Weapon Detection AI Showcased by Vizgard at Security & Policing 2025
Projects
June 29, 2025
Vizgard Selected for BAE Systems Counter-UXV Challenge
Company
February 21, 2025
Security AI Startup Welcomes New CTO to the Management Team
Partnerships
April 18, 2025
Vizgard Joins APPG for Defence Technology
Projects
January 1, 2020
FortifAI AI Sea Trial on Royal Navy’s Uncrewed Submarine
Media
June 2, 2025
Vizgard Interviewed for Janes Magazine Feature
Frequently Asked Questions
FAQ
Most AI demos look impressive in the lab but fail in the field. They don’t scale, add little beyond drawing boxes on a screen, and are too complex for operators to use or maintain.
FortifAI is different. Built for real-world deployment, it scales across cameras, drones and sensors, running multiple ML models in real time and fusing data for reliable performance. It blends advanced machine learning with proven computer vision, supports a wide range of camera and vehicle protocols, and enables autonomous control with seamless integration. With UnifAI providing a tailored ML ops layer for defence and government, systems continuously learn and improve. Already proven across diverse environments, FortifAI delivers reliable, scalable performance from day one.
Partnering with us can save an estimated £3.5M over 3 years and deliver working capability over 50 times faster.
While it’s relatively easy to build a proof of concept using off-the-shelf tools, turning that into a reliable, field-ready system takes serious time, cost, and specialist expertise. FortifAI is already operational, proven, and designed for seamless integration, avoiding the usual delays, risks, and maintenance burden of building from scratch.
FortifAI is designed to integrate with a wide range of defence and security systems, regardless of camera type, deployment method, or environment. It supports both RGB and thermal inputs and has been deployed on unmanned aerial, ground, surface, and subsurface platforms.
It’s also integrated into counter-drone and unmanned traffic management systems, long-range surveillance platforms, and traditional CCTV infrastructure.
Yes. We currently support models from YOLOv7 through to YOLOv12. Using UnifAI, you can convert, optimise, and deploy your own models directly to your FortifAI-enabled devices without needing to share model files or class labels with us.
Yes. FortifAI goes far beyond basic detection and tracking. It can:
Steer multiple uncrewed vehicles and camera types in real time
Fly UAV missions in GPS-denied environments
Geolocate objects and output to systems like ATAK
Enhance visibility with fog removal and image stabilisation
Blur faces and number plates for privacy compliance
Assess threat levels based on behaviour, number plates, and vehicle attributes
Encode and decode metadata in various video formats
FortifAI is licensed per video stream per year, with a minimum of two streams, and tied to the hardware ID of each Jetson or x86 device.
For deployments with 10 or more streams, we offer enterprise plans with reduced per-stream pricing, white-labelling options, and access to trained model files.
Licences are stored locally, so systems remain operational even without internet access.
Yes. FortifAI runs on ARM or x86 computers with NVIDIA GPUs, such as the Jetson range, for real-time edge processing. If your hardware is already in place, FortifAI can be installed and running within an hour.
It integrates via a TCP JSON WebSocket API and has a built-in web interface to support testing. It also supports a growing list of protocols, including ONVIF Profile S, MAVLink, NMEA-0183, Pelco-D, and ATAK, as well as proprietary integrations with platforms like Parrot, DJI, and various camera OEMs.
FortifAI is operational across a range of real-world environments. In the UK, it’s used for counter-drone operations along a flight corridor, at a prison, and by several police departments.
For ground security, it’s active at multiple sites in the USA, an international airport, a coastal facility, and with a NATO specialist end user.
In public safety, it’s installed on CCTV systems covering several major London landmarks.
For drone AI, it’s in use by the UK Ministry of Defence and several UK defence tech SMEs, with systems deployed in Ukraine.
Thank you!
We’ve sent a confirmation email with your access link.
