sinaptik-ai/pandas-ai

PandasAI is a Python library that lets you ask questions about your data using plain English, instead of writing code. The problem it solves: data analysis normally requires knowing programming languages like Python or SQL, which excludes most business users. PandasAI bridges that gap by letting anyone type a question like "What is the average revenue by region?" and get a real answer back. Under the hood, it connects your data, whether it's a CSV file, a SQL database, or a data lake, to a large language model (an AI that understands natural language). When you ask a question, the AI figures out what calculations or lookups are needed, runs them against your actual data, and returns the result. It can also generate charts and graphs on request, and it can work across multiple datasets at once, joining them automatically when needed. For safety, you can run the code execution inside a Docker sandbox, which is an isolated container that prevents any malicious or accidental harm. You would use this when you need to explore or analyze data without writing SQL or Python yourself, for example, a product manager checking sales trends, a founder doing a quick sanity check on numbers, or a developer who wants to skip boilerplate data-wrangling steps. The library is written in Python and works with an external AI model of your choice, such as OpenAI's GPT models, configured through an add-on package.