Enhanced query observability for greater insight and control
Blog post from Vectara
Vectara has introduced a feature called Query Observability to enhance the search experience by providing insights into query history and configurations within its web console and API. This functionality allows users to review past queries, including configuration settings, results, and any errors, to optimize future search queries. Users can adjust various settings, such as the balance between neural-based and keyword-based search, reranking, contextual text display, and language of responses, to tailor the search experience to specific use cases. Query Observability not only logs every query submitted to a corpus, providing transparency and aiding in trust-building with AI systems, but also offers tools for troubleshooting and optimizing configurations. This fosters a better understanding of the system, ensuring that users receive a seamless and valuable search experience.