This postdoctoral fellowship at UNINOVA - Institute for the Development of New Technologies is part of the SMAIS project, focusing on Artificial Intelligence for Municipal Water and Sanitation Services. The position is funded under the Recovery and Resilience Plan (PRR) and aims to develop and implement advanced systems of cybersecurity, artificial intelligence, and data science in public administration, specifically for optimized water management in the Almada Municipality. The successful candidate will engage in requirements engineering, stakeholder needs assessment, data analysis, and machine learning model development for municipal water management. Additional responsibilities include supporting end-user training and public/academic dissemination activities. The work will be conducted at UNINOVA's facilities, with possible activities at the FCT/NOVA Campus and other necessary locations, under the scientific guidance of Prof. José António Barata de Oliveira. The fellowship offers a monthly allowance of €1,851, personal accident insurance, and reimbursement for voluntary social insurance. Applicants must have a PhD in Earth Sciences Engineering (completed within the last three years), proven expertise in data science, machine learning, remote sensing, IoT, and satellite image processing, and at least three years of experience in Portuguese/European research projects. Fluency in English and Portuguese is mandatory, and degrees obtained abroad must be recognized in Portugal. The application process requires submission of a CV, motivation letter, certificate of qualifications or declaration, informed consent statement, copy of ID, and supporting documents, with forms available on the UNINOVA website. The selection process includes evaluation of the CV and knowledge level, with interviews for shortlisted candidates. UNINOVA promotes equal access and non-discrimination in its recruitment. The grant is expected to last eight months, starting in May 2025 and ending by December 31, 2025, with possible renewal within project limits.