Swirl can be tried in Docker and includes four Google Programmable Search Engines to get users started. It also offers Microsoft 365 integration and OAUTH2 support, synchronous or asynchronous search federation via APIs, optional subscribe feature to continuously monitor any search for new results, and many more features. Contributions to Swirl are welcomed and it also offers a managed service, Swirl Cloud. For support, users can join the Swirl Metasearch Community on Slack or email the support team.
Key takeaways:
- Swirl is an open-source search platform built in Python and Django, designed to connect databases, warehouses, search providers, and data siloes without moving, re-indexing or re-permissioning sensitive information.
- It uses AI (Large Language Models) to re-rank unified results from various search engines, databases, noSQL engines, cloud/SaaS services, etc., and supports integration with enterprise services like Microsoft 365, Atlassian Jira, Confluence, JetBrains YouTrack, HubSpot and more.
- Swirl can be tried in Docker, and it includes a detailed guide on how to start Swirl in Docker for MacOS, Linux, or Windows users.
- Swirl offers key features like Microsoft 365 integration, OAUTH2 support, asynchronous search federation via APIs, query transformation support, duplicate detection, re-ranking of results using Cosine Vector Similarity, and optional spell correction using TextBlob among others.