The endeavor brings unprecedented capabilities via a new portable IP component that enables on-device learning for real-time anomaly detection and non-intrusive load monitoring (NILM) at the edge

BOSTON, Nov. 17, 2025 (GLOBE NEWSWIRE) -- MicroEJ, a global leader in embedded software, in partnership with Boon Logic, provider of the world’s fastest AI-based anomaly detection solutions, today announced PicoAI, a new software component designed to bring self-learning artificial intelligence to any smart connected devices, eliminating the need for cloud infrastructure and costly training data collection.

PicoAI is a lightweight architecture optimized for resource-constrained microcontrollers and embedded microprocessors, and doesn’t require any neural network accelerators or GPUs to run. It supports edge-native AI across industries including energy, industrial automation, consumer electronics, and wearables and is powered by Boon Logic’s patented clustering technology, Nano, known for real-time, unsupervised machine learning runtimes. PicoAI enables on-device training and real-time inference for pattern recognition and anomaly detection, without requiring predefined failure cases or cloud resources.

“With PicoAI, we bring real-time learning and anomaly detection capabilities to even the smallest devices,” said Fred Rivard, CEO of MicroEJ. “This marks another step in our strategy to deliver software-defined intelligence at scale, optimized for energy, security and cost.”

Real-Time, High-Density Pattern Recognition and Learning
PicoAI brings powerful self-learning capabilities directly to embedded devices, training on real operational data without relying on the cloud or pre-trained models. By learning what "normal" looks like locally, PicoAI identifies anomalies and usage patterns 1,000 times faster than traditional methods – with precision, even in dynamic and unpredictable environments.

At its core, PicoAI uses an extremely efficient clustering algorithm that supports up to thousands of clusters per model, far outperforming conventional edge AI solutions that typically manage only a few dozen. This makes PicoAI one of the most accurate and scalable pattern recognition engines available for embedded systems. Key benefits include:

• Models train in minutes, not months
• Up to thousands of clusters per model for unparalleled detection accuracy
• On-device intelligence with minimal processing and memory requirements
   

“Introducing PicoAI to the embedded market allows customers to train individualized models in real-time at the edge, which is not achievable with neural network approaches,” said Grant Goris, CEO of Boon Logic. “The integration of Boon Logic’s Nano with MicroEJ VEE brings industry leaders an unprecedented ability to accelerate the delivery of edge native AI across their portfolios.”

Embedded AI at Scale for Energy, Industry, and Wearables
Delivered as a portable runtime integrated into MICROEJ VEE, PicoAI can be reused across devices and hardware platforms without code rewrite. This simplifies integration and shortens development cycles for device manufacturers. PicoAI is available today as part of MicroEJ’s ecosystem of IP components and can be deployed across multiple verticals:

• Energy: DER management, load disaggregation, power quality analytics, NILM
• Factory automation: predictive maintenance, equipment monitoring
• Consumer: adaptive appliances, behavioral modeling
• Wearables: health and fitness monitoring, pattern recognition, anomaly tracking
   

Bringing Self-Learning AI to Every Embedded Device
With MicroEJ, manufacturers can embed and update PicoAI across entire product lines, from energy meters and smart appliances to wearables and industrial control systems.

The new PicoAI solution will be showcased at ENLIT Europe, taking place Nov 18-20 in Bilbao, Spain, in Hall 1 – Booth E32, where MicroEJ will demonstrate real-time DER disaggregation running directly on MicroEJ-powered devices.

About MicroEJ
MicroEJ is a global leader in software solutions for intelligent devices, trusted by industry leaders in consumer electronics, energy, healthcare, and industrial automation. With over 400 million units of MICROEJ VEE sold worldwide, MicroEJ empowers manufacturers to turn standard products into smart, cloud-connected, AI-enabled devices, enabling rapid, complexity-free innovation. By bridging IT and embedded systems, MicroEJ drives scalable transformation that helps industries innovate faster, adapt to market needs, and build a software-driven future for IoT and connected devices.

▷ For more information: Press Kit – Press Room – www.microej.com – LinkedIn

About Boon Logic
Boon Logic’s breakthrough, patented nano-clustering technology for unsupervised machine learning simplifies and accelerates AI development and deployment in diverse application areas such as visual inspection, preventive maintenance, and smart grid. The Boon Nano algorithm performs anomaly detection, model training and clustering 1,000 times faster than traditional modeling methods that are complex and involve lengthy model training that can take several weeks to several months. Boon Nano reduces the need for highly-skilled, hard-to-find AI talent and allows organizations to leverage the value of AI with their existing subject matter experts. It is flexible – autonomously tuning, training, and running in real-time at the edge or in the cloud. For more information visit boonlogic.com. Follow us on LinkedIn and on our blog.

Press contacts:

A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/4efd84af-bb49-4a1c-a6cd-24e4d77fcc16