 Nano-HAI Studio is a search engine for machine learning libraries that does not require any expertise in machine learning or data science. It provides a quick and easy way for any developer to start embedding smart features into any C code and build edge AI-powered IoT devices for applications such as condition monitoring or predictive maintenance. This tool takes as input minimal amounts of data and outputs the best Nano-HAI library for a given use case. Logging this input data is also very straightforward thanks to the data logger feature for the SDWIN sensor tile board. The output Nano-HAI libraries are optimized for microcontrollers where they can both learn and infer independently from the cloud. There are four types of Nano-HAI projects. Anomaly detection, which provides a dynamic model that learns patterns incrementally and infer both directly within the target microcontroller. N-class classification, which is used to detect outliers within data and is especially useful when no examples of abnormal behaviors can be provided on the system. N-class classification, which enables automatic identification of a machine state among many different possible states. And extrapolation, which uses mathematical regression models in order to estimate a target value using other known features. Each project is divided into five steps. First, set general parameters. Then import signal examples that represent the behaviors of the machine to be monitored. Start a benchmark to automatically find the best AI library. Test the best library using the emulator. And finally, deploy the library to the microcontroller. In project settings, define the maximum amounts of RAM and flash memory on the microcontroller to be dedicated to machine learning. Then select a target board and sensor type. All libraries are compatible with any STM32 board with an ARM Cortex-M microcontroller. They are also completely sensor agnostic, meaning that any sensor type or a combination of sensors can be used. In the signal step, import signal examples that will be used as context for the automatic search during benchmark. These signals are raw sensor data that represent the behaviors of the machine or piece of equipment that needs to be monitored. In the benchmark step, the studio uses all input signals provided to automatically search for the best nano-AI library for the use case and optimize it. This library will contain the optimal signal preprocessing algorithm coupled with the best machine learning model and hyperparameters. In the emulator step, the best library found during benchmark can be thoroughly tested in real conditions as if it was running on the target microcontroller. Finally, in the deployment step, the best library is compiled and downloaded ready to be embedded to provide versatile machine learning features with minimal development effort.