Efficient energy use is becoming increasingly significant towards addressing sustainability in the urban environment. Despite rapidly expanding research on energy optimization of buildings, there is still a need for data-driven solutions while managing energy supply and demand. This complex issue is influenced by many factors such as lack of structured data, data privacy concerns, diverse federated data analytics, to name a few. At the same time, building users and managers lack understanding and tools to positively influence its’ performance. Departing from Internet of Things (IoT) approaches, our work aims at narrowing down this gap, leveraging Machine Learning (ML) for pattern recognition and data analytics, towards more informed decision-making on buildings’ energy optimization. Our goal is to join expertise and expand availability of infrastructure, enabling case studies in both Lithuania and Sweden where we will demonstrate the effective application of our tools.