From Connected To Smart

The word Ekkono means Cognition. Using our SDK you can rapidly develop and deploy self-learning, predictive and personalized smart features to your IoT products. We make connected things smart.

Ekkono’s strength is that we actually do machine learning at the edge. Our models can learn how each individual unit of your product is used in super-local conditions. Moreover, we can do this on microcontrollers in really small devices. For example, we can enhance your product with predictive and condition based maintenance. Or how about auto-configuration and auto-installation. Implement virtual sensors or smart battery management? With Ekkono, you can turn all those buzz-words into reality!

The product

The Ekkono SDK is an embedded software library built for developers. We add incremental learning with integrated pipelined data processing for real-time continuous learning on your device. Our toolbox includes change and anomaly detection, signal processing and data validation with conformal prediction. Furthermore, we supply hot swappable models with a very small footprint, and with no third-party dependencies.

Get to know the product.

We are pioneering the future of Federated Learning with the prestigious support of the European Innovation Council (EIC). Discover more about our game-changing Synthesis product here.


 1  #include "crystal/crystal.h"
 2  #include "crystal/portable.h"
 3  ek_status_t status;
 4  ek_model_handle_t model;
 5
 6  unsigned char mlp_model_data[] = {0x01, 0x00, 0xb4, 0x01, ..., 0xf1, 0x94};
 7  float instance[] = {1.07f, -0.32f, 209.22f, 0.00f};
 8
 9  model = ek_model_load(mlp_model_data, &status);
10  float prediction = ek_predict_and_train(model, instance);

About Ekkono

Ekkono is a Swedish software company founded in 2016. The headquarters are located in Träslövsläge, on the west coast of Sweden. However, our diverse team of experienced entrepreneurs, Machine Learning Engineers and Data Scientists are scattered around the world.

Trusted by

Sustainability

The world is changing rapidly around us. We believe that with innovation and technology, there are enough resources on earth for everyone. For instance, smart IoT devices will:

Extend product life with predictive maintenance
Decrease pollution by automated engine optimization
Harvest new functionality from old hardware

Get in touch

Do you want to enhance your products with machine learning? Or do you want to know more about machine learning in general? For instance, you might have some thoughts around conformal prediction? Moreover, you may wonder how to detect failure in a fail proof environment?

We are happy to help! Just get in touch.

Technical introduction

Join us

We find pleasure in discussing different aspects of statistics. To dive into problems together. To celebrate our victories. We enjoy talking about nerdy research, dissecting algorithms and finding nuggets of order in chaotic streams of data. We believe that the world would be a better place if the devices were smarter, and we believe that we can make them smarter.

What do you believe? Join us.

Newsletter sign up

Sign up to our newsletter! This will give you access to get the latest news, updates and insights from Ekkono. You will receive interesting articles on edge machine learning related topics, technology, research and projects.