Qux360

Qux360 is an experimental Python library for AI-assisted qualitative analysis. Its purpose is to enhance both the quality and speed of qualitative analysis by providing features that support researchers at various stages of the qualitative analysis process.

This includes ensuring privacy and a secure environment to protect intellectual property (IP) information before beginning the analysis. The library allows for both bottom-up and top-down approaches to topic extraction and insight generation, while also emphasizing validation and transparency in AI outputs to maintain quality and promote trust. Validation is a core principle in Qux360; every use of large language models is designed to be transparent, explainable, and open to scrutiny.

The goal is to assist developers in creating trustworthy, interactive qualitative analysis experiences without compromising flexibility in applying Qux360’s built-in quality assurance mechanisms

Qux360 is built on Mellea, a generative computing library that offers robust and validated prompting techniques.

Currently, the library has been tested exclusively with the meta-llama/llama-3-3-70b-instruct model; other models may require prompt adjustments.