Nicola Piovesan
HUAEI Paris Research Center, France
Title: From Language Models to Autonomous Agents: AI for Intelligent Telecom Networks

Abstract:
As telecom networks scale and grow more complex, traditional operation and management relying heavily on human expertise can no longer meet modern efficiency and adaptability demands. While large language models (LLMs) have started to empower network operations such as procedure assistance and root-cause analysis, directly applying standalone LLMs remains limited in dynamic, real-world telecom environments. These challenges call for a new paradigm: agentic AI systems that go beyond basic LLMs. By integrating LLM reasoning with planning, tool usage, reflection, memory, code execution, and network world models, agentic AI evolves from passive assistants to active decision-makers. This talk discusses how agentic AI enables autonomous telecom operations—with AI agents continuously monitoring, diagnosing, and optimizing networks with minimal human intervention. We focus on agent architectures that embed domain knowledge and network world models for reliable, interpretable, and adaptive decision-making in complex networks.

Short Bio:
Nicola Piovesan is a Senior Researcher at Huawei Technologies in Paris, focusing on research at the intersection of artificial intelligence and future networks. He received his Ph.D. in Network Engineering from the Polytechnic University of Catalonia (UPC) in 2020, and holds B.Sc. and M.Sc. degrees from the University of Padova, Italy. His research centers on generative AI, autonomous agents, and large-scale network modeling for next-generation telecommunications. He has published over 50 papers and holds more than 10 patents, and has been awarded the Huawei GTS President Award (2021) and Huawei Quality Star Award (2024). He also collaborates widely on global projects for AI and energy efficiency in next-generation networks.