Data2Value
by LEONARDO S.P.A.
Suite for data valorization and for the creation and governance of AI models
Data2Value is designed to address the challenges of integrated Big Data analytics and AI governance. It's integrates modern open-source technologies and proprietary modules, designed to ensure security, scalability and interoperability environments.
It is include two inter-operating but independent products:
Data Platform: a set of integrated tools to ingest, manage, valorize, enriche and presente large volumes of data, with particular attention to the highest standards of security (data centric security, data in use, data governance).
Main components:
- Data Lake: a ready-to-use multi tenant secure platform for ingesting, storing and retrieving structured, semi-structured and unstructured data of any size
- Data Processing and Presentation: provides functions for real time and batch elaboration leveraging in-memory execution engine and horizontal scalability and parallel processing. Publish-subscribe capabilities, event management, scheduling and workload management as well as visual ETL (Extract Transform Load) and a Secure Query Engine further empower data processing capabilities. The presentation provides a module for visualizing and interacting with data through graphical data quality interfaces and business intelligence tools: tables, charts and other types of representations.
- Data Security and Governance: implements a single, secure and centralized reference point for data control. By leveraging search and discovery tools and connectors to extract metadata from any data source, it simplifies data lineage, data protection, analysis, and pipeline management, as well as accelerates ETL processes.
AI Engine: a platform to manage and govern AI/ML models in terms of: Responsible AI, Repository (Model/Dataset), Serving, Active Monitoring, Experience Collector and Benchmark Evaluation.
Main components:
- AI model build: set of tools and best practices aimed at developing and maintaining Machine Learning (ML) models reliably and efficiently. It includes the following features: Model Training, Model Tracking, Benchmark.
- Gen AI Engine: the system ensures accuracy, maintaining context and ensuring ethical and regulatory compliance through a processing engine, an orchestration framework, an information retrieval and enrichment pipeline, and monitoring. It includes the following features: Cognitive Endpoints, AI Pipeline, Agentic Framework, Feedback, Model Repository & Marketplace, AI Serving, AI Monitoring.
- Secure & Responsible AI: provides practices and tools to ensure safety, fairness, transparency and compliance with ethical principles. It includes the following features: Responsible AI, Red teaming, Confidential AI, AI Brain Guard
