 Hello, are you looking for a simple, fast and affordable way to optimize your design with HJI? Generally, creating an HJI device like a predictive maintenance application comes with a lot of challenges. For example, in a lot of companies there might not be qualified datasets available or maybe no data scientists. Very often in existing AI applications there is a strong cloud dependency and this might lead to huge hardware requirements like processing power, power consumption and memory footprint. And in the end, it often means that these applications come with a quite significant upfront investment. All of this means that a lot of companies have studied the subject but don't really have an HJI product on the market yet, but honestly, we've come up with a solution for you. Nano HJI Studio is a unique solution that's answering all those challenges. It allows you to create a state-of-the-art AI solution in a simple, fast and affordable way. The way it works is that you can choose the nature and the quantity of data that you log for your input dataset and then Nano HJI Studio will automatically determine the best combination of machine learning models based on your input dataset and on your application. In fact, we've already deployed this solution at various customers and their feedback was that it's very easy to use tool and that they could do it without an AI expert. As an example, here we can detect abnormal behavior of a motor simply reading a current sensor. We're able to do the learning directly on the target so that no cloud learning is needed. It will result in a low footprint in terms of memory and microcontroller processing power. The library will then automatically confirm that everything is running fine in normal condition here in green and as soon as something goes wrong, it informs you that you are in anomaly here in red. Let's now consider a new stage as a normal stage simply by starting a new learning. This is a unique key of Nano HJI Studio, which is on-device learning. It means you can directly retrain the model on the microcontroller itself, avoiding interaction with the cloud or a PC. This saves power and time. Now the learning has been completed and the new algorithm understands this stage as a new normal stage as you can see in green. Nano HJI Studio addresses all those HJI challenges on a very easy, quick and affordable way. If you want to know more about this product, please contact your local ST representative or you can visit our dedicated product webpage on ST.com.