I need a small-to-mid-size Python 3.x project entirely centred on data processing. The script (or self-contained package) should be able to ingest a raw dataset from a file path or stream, apply a clear transformation pipeline that I can modify later, and write the cleaned, structured results back out. The exact data format is still flexible—I’ll provide sample files once we start—so please write modular code that separates the I/O layer from the core transformation logic. Pandas, NumPy, or similar libraries are welcome as long as they are the right tool for the job and you keep external dependencies to a minimum. My priorities are: • Readable, well-commented code that follows PEP 8 • A concise README explaining setup, usage, and how to extend or swap in other data sources • At least one unit test demonstrating that the main processing routine works on sample input The focus is processing; a fancy user interface or external API calls are not required, though you may include lightweight CLI argument parsing if that makes operation smoother. Hand over the finished Python files, any requirements.txt or environment.yml, the README, and the test suite. Once everything runs cleanly on my machine with a single command, the job is complete.