
1. Technology overview
Plumerai is a London-headquartered company with an office in Amsterdam, led by founder and CEO Roeland Nusselder. It licenses AI software to camera makers so that home, business and retail cameras can recognise people, vehicles and events on the device itself, even on battery-powered cameras with inexpensive chips.
Its Tiny AI models run on the camera's own processor, including Arm Cortex-M, Cortex-A and RISC-V chips, and the company says they work on chips costing about 1 USD with 1 MB of memory. Features include people, vehicle, animal and package detection, advanced motion detection, familiar face and stranger identification and multi-camera re-identification. A video intelligence pipeline links this on-camera AI with vision language models in the cloud, which are activated only when needed for AI Video Search in natural language and AI Captions. Chamberlain Group, maker of myQ and LiftMaster, is a named customer, and chip partners include Ambarella, Realtek, Ingenic, SigmaStar, Renesas, ST, NXP and Qualcomm.
2. Highlighted features & specifications
- Object detection: on-device detection of people, vehicles, animals and packages.
- Face identification: familiar face and stranger identification to filter alerts.
- Multi-camera re-identification: follows the same person across several cameras.
- AI Video Search: natural-language search across camera video history using vision language models.
- AI Captions: automatic descriptions of what happens in each clip for richer notifications.
- Plumerai Inference Engine: runs AI models efficiently on CPUs and microcontrollers.
3. Potential application
A Vietnamese camera brand can license Plumerai software to add person and package detection to low-cost Wi-Fi cameras. A retail chain can use multi-camera re-identification and video search to review incidents across stores. A building or apartment security provider can cut false alarms by identifying familiar residents and staff.
4. Benefits
On-camera detection reduces false alerts and the volume of video sent to the cloud, lowering bandwidth and storage costs. Plumerai reports customer benchmarks where AI Video Search was more accurate than cloud-only options at up to 135 times lower cloud cost. Running on low-cost chips lets camera makers add AI features without a hardware redesign. Familiar face identification cuts unnecessary notifications for end users.
5. Market trends & opportunities
Smart camera makers compete on AI features but face pressure on chip cost, battery life and cloud bills. Combining tiny on-device models with selective use of vision language models is becoming a common design. Asia hosts most camera chip vendors and manufacturers, and Vietnamese and Southeast Asian camera brands serving homes and shops are looking to add such features to affordable devices.
6. Other information
Headquartered in London with an office in Amsterdam; founder and CEO is Roeland Nusselder. Raised an 8.7 million USD Series A in September 2025 led by Partech and OTB Ventures, with Acclimate Ventures, bringing total funding above 17 million USD. Its Tiny AI runs on millions of cameras, and the company is expanding from home security into enterprise security and retail.
One platform connecting startups, corporations, research institutions, regions and investors











Playbook & Handbook · February 6, 2026
Winning Accelerator Playbook 2026 by BambuUP
EDTECH: LEVERAGING AI-POWERED TECHNOLOGIES TO TRANSFORM THE LEARNING EXPERIENCE








































