 With Industry 4.0, we address a deep transformation of the manufacturing industry where data science and its applications bring breakthrough performance in terms of profitable sustainability of the production plant. At its core, we see data become digital and interoperable. High-speed data logging capabilities enable data aggregation and then big data, which are instrumental to the creation of advanced solutions based on artificial intelligence and machine learning. This way, events captured in the production plant can be automatically labeled and classified, and predictive algorithms enable us to optimize overall process and performance, while they may even enable us to predict potential problems, thus eliminating machine downtime. The STWIN platform and its software ecosystem have been designed in order to support and facilitate this evolution. The STWIN, short for Sensor Tile Wireless Industrial North, is a small and compact 5x5 centimeter board which integrates industrial grade and high-quality sensors capable to measure parameters such as inertial, environmental, vibration, acoustics and even ultrasound. It comes in the development kit along with battery, plastic holder and programming cable. The STWIN 1B just announced features an ultra-low-power STM32 L4 Plus, well-known in the IoT developers community for its flexibility and optimal trade-off between high performance and low power consumption. Wireless communication is supported via a Bluetooth radio powered by our Blue Energy Dash Tuning. ST Safe Secure Element provides a cost-effective and certified support for end-to-end security from data storage to device authentication. Vibration analysis can be performed inertially at up to 6 kilohertz and acoustically in the ultrasound range at up to 80 kilohertz. STWIN developers may leverage on a wide software ecosystem that includes example of artificial intelligence machine learning and industrial solutions supported by the STM32 community. Now let's put all of these in the context of data science. There's two key operations related to data that must always be part of a data science project. You always must get the data and be able to use the data. In order to capture the data you may place the STWIN on an engine, a gear, a pump or a gas pipe just to mention a few examples. You may want to control the whole acquisition process with a dedicated smartphone app and store the dataset on a host be it a PC or a cloud-based. Sensor data we would like them to be fully synchronized and properly labeled so that in a host environment they can be directly used to feed our preferred algorithm generation toolchain. All these operations are supported by the high-speed data logging framework that comes with the STWIN. In the high-speed data logger the STM32 firmware enables us to capture all the onboard sensors even when they run all simultaneously at their maximum sampling rate and to stream them over USB or to write them on the onboard SD card. The overall data rate could be up to 6 megabit per second. The acquisition process can be controlled via a dedicated smartphone app through which we can select which sensor we want to log or we can define the values of key parameters of those selected sensors including the machine learning core, start and stop the acquisition and even label the data in real time during the logging session. The same set of operations can also be performed with the board connected via USB to a host directly from within a Python environment. In this way data scientists can have full control of the acquisition process without having to deal with writing a single line of low-level code. Data sets captured are fully interoperable with the ST and third-party AI and machine learning toolchains including any Python-based development environment. Consider for example feeding the data sets to NanoEdge AI studio. The tool generates an STM32 machine learning library ready to be integrated in the dedicated software example available at ST.com. This way you can deploy your STM32 based solution directly on the field where it will autonomously learn to monitor and detect anomalies. All the information on boards and software are available at our website at www.st.com. Please do not hesitate to contact your local ST sales representative for further information. Thank you.