pandas-dev/pandas ? reverse-engineered prompt

Build me a powerful Python library for data analysis and manipulation, let's call it pandas. Its main purpose is to make working with structured data, like you'd see in a spreadsheet or a database table, really easy and fast.

The library should be able to load data from common file types like CSV and Excel, and it needs to be really good at cleaning up messy, real-world data, especially handling missing values automatically.

I want it to have powerful features for data operations. Users should be able to easily group data to run calculations (like finding totals per category), merge different tables together, and pivot data. It also needs strong support for working with time-series data, like stock prices or sensor readings. Essentially, it should be the fundamental building block for anyone doing practical data science in Python.