I have a large collection of raw text documents and images that I need turned into clear, actionable insights. The job centres on exploratory data analysis rather than predictive modelling: I want to understand patterns, themes and relationships hidden inside both formats. Here is what I expect at the end of the engagement: • A well-commented Python (or R) notebook that walks through preprocessing, cleaning and exploratory techniques applied to the text and image sets. • Concise visual summaries—charts for the text, heat-maps or feature plots for the images—exported in high-resolution formats I can drop straight into presentations. • A short written report (PDF or Markdown) highlighting key findings, unusual correlations and any recommendations that emerge from the analysis. Feel free to lean on standard libraries such as pandas, NumPy, NLTK/spaCy, OpenCV or PyTorch-based utilities—whatever helps you surface meaningful insights quickly and reproducibly. If you have other preferred tools that suit mixed unstructured data, I am open to them so long as the final deliverables remain easily reusable. I’ll supply a structured folder containing sub-directories for /text and /images along with a brief data dictionary. Please keep all code modular and reproducible so I can rerun it when new files arrive. Once I review the notebook, visuals and report and they run cleanly on my machine, we are done.