Pinned Loading

1. pyhound pyhound Public

   Hunt down retrieval problems in RAG/LLM systems. Hybrid retrieval diagnostics for embedding quality, vector search, BM25, and reranker analysis. MIT licence.

   Python

2. statguard statguard Public

   Python library for data quality and validation — declarative contract DSL, schema checks, drift detection (PSI + KS), and anomaly detection across Delta Lake, Iceberg, Parquet, Avro, and more. 13x …

   Rust

3. ClusterAudienceKit ClusterAudienceKit Public

   Python library for customer segmentation in Martech pipelines — RFM analysis, clustering, streaming updates, and drift detection in a single pip install.

   Rust

4. StreamXL StreamXL Public

   Streaming XLSX reader for Python powered by Rust. Reads large Excel files row-by-row at constant memory. 4-5x faster than openpyxl. PyO3 bindings, pip and uv installable.

   Rust

5. prismnote prismnote Public

   Modern open-source data science notebook with Rust backend, Python execution, and AI assistance

   Rust