With the new generation of iFinder and the new iAssistant, IntraFind offers AI-based tools that optimally support employees in their daily work and make organizations more efficient. The iFinder enterprise search software speeds up the search for relevant information across a wide range of data sources, while the iAssistant extends the iFinder with AI-generated answers and summaries from the organization’s own data.
Enhanced search quality with LLMs
The new generation of iFinder optimizes search quality by utilizing Large Language Models (LLMs), which enable semantic vector search. This enables the enterprise search software to identify even better matching documents or text passages that take into account the search query, thereby further increasing the relevance of the search results.
Chat with your organization’s own documents - the iAssistant
Together with the new generation of iFinder, IntraFind is launching the new iAssistant. This tool uses a generative language model to provide users with an alternative approach to information provision. They receive answers to questions including logical follow-up questions as well as information on the sources from which the answers were generated. This allows them to verify the correctness of the answers at any time - a basic prerequisite to give users confidence in the capabilities of generative AI.
Employees can also use the iAssistant to create summaries from the organization’s own documents and chat with them. To obtain the information they need, they can ask questions about documents and obtain further details by asking questions. The language model used is multi-lingual, so that, for example, a question asked in English is answered in English based on German-language information.
Technically, the iAssistant is based on a RAG architecture (Retrieval Augmented Generation). In the case of a query, the company data checked for access rights is first searched for relevant information and passages in the same way as a classic search. These are then passed to the language model to generate the answer. This methodology ensures that GenAI can be used securely with the organization’s own data. End users only receive information in their responses for which they are authorized.
The iAssistant is available as an extension module of the iFinder, but it can also be used as a stand-alone application and can be seamlessly integrated into a digital workplace, the intranet, a wiki, a portal or other systems. If companies are already using a chatbot, they can continue to use its user interface and have the iAssistant provide the answers to the questions in the background.
Further innovations optimize access to relevant information
Numerous other iFinder innovations also improve access to relevant information. The redesigned search interface, for example, anticipates users' needs and personalizes their searches.
In general, IntraFind is introducing extensive changes to both the front-end and back-end architecture with the new product generation. These include a switch from Elasticsearch to OpenSearch. Overall, the innovations increase the performance of the software, improve its customization options and facilitate updates to future new versions.
‘The iFinder speeds up the search for relevant information, even in large amounts of data, and the iAssistant provides precise, contextualized answers,’ explains IntraFind CEO Franz Kögl. ‘This opens up entirely new possibilities for companies in terms of saving and accessing knowledge. The great thing is that enterprise search is increasingly recognized as the essential technical foundation for effectively using generative AI in companies — even at the management level. Our customers are also closing the gap between ChatGPT enthusiasm and the concrete benefits of the technology in the company. With iFinder, we enable comprehensive pre-processing for modern LLM-based AI, allowing us to quickly and easily implement our customers' innovative application ideas.
