Skip to main contentSkip to footer
Projects

FIDEA – Intelligent Fabric for Knowledge Diffusion and Exploitation with AI Agents

Turning Expert Knowledge into Action Through Artificial Intelligence Agents
Funding: IFAB call for members
Enabling technology: Artificial Intelligence, Knowledge Graph, LLM, Machine Learning

Status: In corso

Sustainable Development Goals

FIDEA is an application platform designed to transform the tacit knowledge of company experts into a structured, searchable and continuously updated digital asset. Through a system of AI-powered cognitive agents, the platform enables organisations to formalise internal know-how, make it accessible in real time through conversational interfaces and preserve it over the long term within corporate knowledge graphs.

Delivered as a Software-as-a-Service solution, FIDEA integrates domain-optimised small language models, fine-tuned for specific use cases, together with short- and long-term memory and automated input collection engines. Designed to adapt to different industries, FIDEA is primarily intended for small and medium-sized enterprises seeking to enhance the value of their internal knowledge assets and gain access to the capabilities of IFAB’s AI Factory.

Objectives

The project aims to develop a SaaS platform enabling SMEs to access the resources of IFAB’s AI Factory for the management and enhancement of internal knowledge. Its main objective is to build and orchestrate a network of cognitive agents capable of formalising the know-how of company experts, updating it in real time through conversational interactions and making it searchable in natural language.

The project also aims to demonstrate the model’s replicability across different application domains, positioning FIDEA as a horizontal solution that supports the competitiveness and efficiency of organisations adopting artificial intelligence.

Initial Challenge

Every year, organisations risk losing a body of expertise that is extremely difficult to recover. The tacit knowledge held by experts, technicians and professionals is rarely formalised systematically and often disappears as staff change or business processes evolve.

This challenge is compounded by the lack of tools capable of dynamically updating such knowledge and making it accessible through natural-language queries within day-to-day operational contexts.

Solution

FIDEA addresses this challenge through an orchestrated network of cognitive AI agents capable of interacting with company experts, extracting their knowledge through structured conversational sessions and encoding it within a persistent knowledge graph.

Fine-tuned language models, combined with semantic reasoning capabilities, provide accurate and context-aware responses, including in domains characterised by highly specialised terminology. The platform’s modular architecture supports progressive customisation, allowing the system to adapt to the needs, processes and workflows of each organisation.

Benefits

FIDEA enables SMEs to preserve and enhance their intellectual capital, reducing operational errors, accelerating the onboarding of new employees and improving compliance with complex regulatory procedures.

The model is inherently replicable across multiple application domains, from manufacturing to professional services, and provides a scalable foundation for developing specialised AI assistants that enhance organisational competitiveness and efficiency.

IFAB’s Role

IFAB participated in the project as a funding organisation, selecting it as part of its investment programmes in research, innovation and technology transfer.

IFAB’s contribution consisted of recognising the project’s scientific and practical value and providing financial support, thereby enabling the technical partners to carry out the planned research and development activities.

This involvement further confirms IFAB’s mission as an accelerator of high-impact initiatives for the local area and the wider production ecosystem.

For further information, please contact: projects@ifabfoundation.org

Other projects in the field of Cross-cutting Technologies