Home

Qux360

An experimental Python library for AI-assisted qualitative analysis

Validation is a first-class concept in Qux360 — every use of large language models is designed to be transparent, explainable, and open to scrutiny. The goal is to help developers build trustworthy, interactive qualitative analysis experiences while retaining flexibility in how they apply Qux360’s built-in quality assurance mechanisms.

Qux360 is built on Mellea, a generative computing library that provides robust and validated prompting techniques. The current Qux360 version supports interview data with a single participant (interviewee) only, with plans to expand this scope in future releases.

Key capabilities

• Import interview transcripts in DOCX, XLSX, or CSV formats
• Export processed transcripts in XLSX or CSV formats
• Speaker anonymization
• Statement anonymization using local, privacy-preserving entity detection
• AI-assisted interviewee detection
• AI-assisted top-down topic extraction
• Bulk processing across collections of interviews (e.g., anonymization)

📜 License

Qux360 is licensed under the Apache License 2.0.