Introducing Search Toolkit

Search Toolkit is an open source framework that combines data ingestion, retrieval, and evaluation into a single system for building search pipelines in AI applications. Teams building search infrastructure currently spend significant engineering time assembling separate tools with different interfaces and data assumptions, and this toolkit aims to reduce that integration work by providing a shared interface across all three components. The framework supports multiple retrieval methods including sparse retrieval and dense embedding-based retrieval, works across cloud and on-premises environments, and includes built-in evaluation metrics to measure search quality independently from generation quality. This matters particularly for enterprises managing multiple search problems across different data sources and for AI agents that need to access enterprise context at scale.