Building ultra-tiny ML models for any wireless SoC with Neuton

The tiny-ML experts in Neuton have developed a unique neural network framework, not based on any existing technologies such as TensorFlow, PyTorch or similar, to create ultra-tiny ML models. This framework allows building models 10x smaller than competing frameworks, down to as small as single-digit kilobytes, with great generalizing capabilities. This gives the models the same or better accuracy and faster inferencing than these competing frameworks, resulting in lower power consumption. As of 2025, Neuton is now a part of Nordic Semiconductor.

In this webinar, Danil from the Neuton team will lead you through the workflows of creating ultra-lightweight and ultra-fast Neuton models to enable Edge AI capabilities on any SoC. Learn how to collect data to prepare a dataset and build a highly optimized Neuton model that can run on everything from the flagship nRF54L15 SoC to the most constrained devices in the Nordic lineup, like the nRF52805 SoC. After the hands-on tutorial, you will be able to ask questions directly to the team behind developing the technology that made it possible in a live Q&A session.

Agenda:
00:00 Intro and hardware overview
03:00 Demo: gesture remote control
06:30 Collecting and preparing a dataset
25:40 Building AI models with Neuton’s online tool
47:05 Q&A with the Neuton developer team

This is a recording of one of two live sessions held on June 25th 2025, with the Q&A part of both session appended at the end. The slides from the webinar can be downloaded here: https://view-su3.highspot.com/viewer/...
The repository for the Thingy:53 sample project is found here: https://github.com/Neuton-tinyML/neut...