Our AI Agents are built on a Retrieval-Augmented Generation (RAG) architecture that combines retrieval capabilities with generative AI. They use Microsoft's Azure OpenAI Service, allowing the model to access real-time data from your various internal sources—such as SharePoint, OneDrive, and custom databases—through a sophisticated search and indexing mechanism.
The architecture is designed for ease of deployment, with built-in connectors to Microsoft 365 environments, and it supports flexible customization, enabling fine-tuning and bespoke additions to the model with specific data and access controls. The Agent delivers accurate, contextually relevant insights tailored to the specific needs of your organization.
-
The Agent ingests and automates the indexation data from internal databases, sharepoints, email, shared drives, and external sources (e.g., social media, news feeds).
It also has self-service capabilities that allow users to easily upload custom folders, PDFs, and documents for on-the-fly ingestion.
-
The Agents employ Large Language Models (LLMs) to quickly summarise or clarify relevant data segments.
It also has voice-to-text queries that let teams ask spontaneous questions during meetings, grounding discussions in real facts.
-
Every piece of AI-generated output includes citations to the relevant source documentation, ensuring transparency and trustworthiness.
Users can verify the original source of any information used in the generated response.
-
A private repository stores chat sessions, which can be used to train the model on frequent or domain-specific questions, refining accuracy over time for enterprise users.
The Agent also has end-to-end encryption and role-based access control protect sensitive information.