Mattia Redaelli

Mattia Redaelli is an AI engineer whose work at Quantyca addresses a practical challenge for large organizations: making enterprise AI agents discoverable, governable and reliable across distributed teams. He helped develop Amplifon’s AmplifAI infrastructure, combining registries for agents, tools and business use cases with reusable development blueprints.

Redaelli earned a master’s degree in computer science and engineering from Politecnico di Milano before joining Quantyca. His graduate research examined load balancing and scalability in Kabis, an Apache Kafka-based framework for Byzantine fault-tolerant distributed systems. He subsequently investigated knowledge graphs for organizational knowledge with Quantyca colleague Mauro Luchetti, exploring how explicit relationships make dispersed enterprise information usable by intelligent systems.

For Amplifon, Redaelli worked alongside Luchetti and Amplifon’s Sonny Merla on an enterprise agent-governance platform. Its core contributions include:

• Enterprise MCP and A2A registries: Catalogs connect Model Context Protocol servers and Agent-to-Agent services with ownership, authentication requirements, agent cards and associated business use cases.
• Use-case lineage and impact analysis: Visual dependency maps connect applications with their agents, tools and models, helping teams identify systems affected by component failures or changes.
• Framework-agnostic agent blueprints: Shared templates provide standardized interfaces, authentication, containerization, cost tracking and Langfuse observability while preserving each team’s choice of agent framework.
• Self-documenting deployment pipelines: GitHub Actions publishes agent cards and MCP server metadata into shared registries alongside application releases, keeping governance records synchronized with deployed services.

The platform gives enterprise developers operational guardrails and reusable infrastructure without requiring every team to adopt the same implementation stack.