“WAISE” PROJECT
Have you ever wondered how an AI-powered chatbot can transform your collaborative projects by leveraging the power of a flexible wiki and a traditional database?
Learn how WAISE, part of the NGI Search initiative, uses advanced AI and large language models to enhance knowledge sharing by offering a unique blend of innovation and practicality.
Are you keen on learning more? Join us and see how WAISE is set to revolutionise the world of open-source collaboration.
Can you introduce yourself and your project?
Hello everyone!
We are excited to share our latest project, WAISE (Wiki AI Search Engine), which is being developed as part of the NGI Search initiative. At XWiki SaS, we have always been passionate about empowering collaboration and knowledge sharing through our open-source wiki platform. With WAISE, we are taking this mission to the next level by leveraging the power of AI and large language models (LLMs).
XWiki is not just another wiki software—it’s a fully featured and extensible platform.
We designed XWiki to provide the building blocks you need to bring your collaborative projects to life (intranet, project management tool, etc.). Our unique “second-generation” wiki approach combines the flexibility of a wiki with the structure and power of a traditional database application.
What makes XWiki stand out? Here are just a few examples:
- – App Within Minutes:
Create simple applications directly from the wiki interface without coding - – Structured Data:
Store and manipulate structured information using classes and objects - – Advanced Permission System:
Manage fine-grained access control for spaces, pages, and content - – Scripting Support:
Build advanced functionality and macros using Velocity and Groovy scripting - – Extensibility:
Customise with hundreds of available extensions.
What services or products do you offer?
With the WAISE project, we are building an application server that provides an LLMs-powered chatbot. The key innovation is that this chatbot can be integrated into any application to answer questions based on the content and data in that application while respecting the current user’s permissions.
In developing WAISE, we are taking full advantage of XWiki’s powerful features. For example, we use the App Within Minutes functionality to rapidly prototype and iterate on the WAISE application directly from the wiki interface. This allows us to quickly build out the core components like the user interfaces for managing language models or collections of documents.
We also leverage XWiki’s support for structured data and integration with Solr 9 to efficiently index and query the content that powers WAISE’s AI-assisted search. The advanced permission system ensures that WAISE respects user access controls when retrieving information.
By building on top of XWiki, we get a robust and flexible foundation for WAISE that allows us to focus on AI and search innovation.
Some highlights of the WAISE architecture:
- – Indexing of content from external applications via a REST API
- – Retrieval of relevant context to augment the user’s question
- – Authentication support to embed the chat securely in other apps
- – Planned integration with XWiki itself to directly index wiki content
What milestones have you achieved so far since your project launch?
Since the launch of the WAISE project, we have made significant progress:
Implemented support for indexing content from external applications using a REST API
Developed retrieval capabilities to find relevant context for a user’s question
Added authentication mechanisms to securely embed the WAISE chat in other applications
What have you achieved with your idea thanks to the NGI Search project?
The NGI Search funding has been instrumental in allowing us to dedicate development resources to the WAISE project. With this support, we were able to progress quickly and comprehensively.
What are your goals for the middle/long-time future?
Looking ahead, we have some exciting plans for WAISE:
- – Implement an integration with an external open-source application like OpenProject to showcase the power of the
WAISE chatbot in enhancing existing tools. - – Deeply integrate WAISE with XWiki, indexing content directly and allowing users to reference wiki pages explicitly as
context for their questions. - – Continue to refine and optimise the LLM prompting and retrieval to provide the most relevant and accurate answers.
Do you have any advice for those who are looking for public funding?
For other open-source projects and companies seeking public funding, our advice would be:
- – Align your project goals with the strategic objectives of your target funding programs. The NGI Search’s focus on privacy-
preserving, trustworthy search and discovery was a great fit. - – Collaborate with other open-source projects and communities. We believe strongly in the power of open-source
partnerships, as shown by our plans to integrate with OpenProject. - – Think big, but have a concrete plan. Articulate your vision but back it up with a detailed architecture and roadmap, as we
have tried to do with WAISE.
We are grateful to the NGI Search program for its support and are excited to continue working with WAISE to bring the power of AI-assisted search and discovery to the open-source world.
