I have several structured datasets—mostly spreadsheets and database exports—that need to move from raw form to decision-ready insights. Your task is to build an efficient AI-powered pipeline that first cleans and preprocesses the data and then presents the key findings through clear, compelling visualisations. You’re free to choose the stack that suits you best; whether that’s TensorFlow, PyTorch, scikit-learn, or another reliable toolset, I only ask that the code is well-documented and reproducible. Throughout the engagement I value transparent progress updates and quick responses to questions, so smooth communication is essential. Deliverables I expect: • A fully cleaned and preprocessed version of each dataset • Reusable scripts or notebooks that automate the cleaning steps • Insightful visualisations (dashboards or static charts) that highlight trends, outliers, and actionable metrics • A brief walkthrough or read-me explaining how to rerun the pipeline If you can combine rigor with creativity and hit deadlines without sacrificing quality, let’s get started.