The professional home for the engineering and technology community worldwide.

Event

Date

Apr 28 2026

Time

6:00 pm - 8:30 pm

Location

Wentworth Institute of Technology, Beatty Room 426 (Building: Beatty Hall)
550 Huntington Ave, Boston, MA 02115

Automation: A Practical Walkthrough to Agentic AI Framework (Semantic Kernel) and its Applications in Power and Energy Systems

Power & Energy Society

Speaker: Barsha Upadhyaya, AI Analyst, First Solar

Time: Refreshments start at 6:00pm; Talk begins at 6:30pm

Location: Wentworth Institute of Technology, Beatty Room 426 (Building: Beatty Hall)

550 Huntington Ave, Boston, MA 02115.  (Free Parking at West Parking Lot)

The rapid advancement of artificial intelligence is reshaping industries at an unprecedented pace. Beyond traditional automation, a new paradigm is emerging – one where AI systems do not simply execute predefined tasks, but reason, plan, and act autonomously in response to complex, dynamic environments. Agentic systems hold significant promise across a wide spectrum of industries — including power and energy, manufacturing, supply chain, IT, and many more — extending well beyond conventional rule-based or purely data-driven approaches. In this session, we will explore one such agentic framework called Semantic Kernel, showcasing its capability to run a reasoning-based, fully autonomous fault analysis in a solar panel manufacturing environment – creating what we call a centralized reasoning agent.

Speaker’s Bio:

Barsha Upadhyaya is an AI Analyst at First Solar, where she works on applying large language models and agentic AI systems to manufacturing operations in the solar energy sector. She holds a Master’s degree in Computer Science Engineering from the University of Toledo and has experience spanning applied machine learning, agent-based system design, and production-scale AI deployment. At First Solar, her work focuses on building Semantic Kernel–based reasoning agents and centralized AI platforms that support incident response, operational visibility, and decision-making in photovoltaic manufacturing. Barsha is also the author of a peer-reviewed paper published at the NAPS 2025 Conference on using LLMs and Semantic Kernel agents for near–closed-loop control in PV manufacturing.

Free and Open to the Public; RSVP is appreciated

Visit our web page to RSVP and for further details – IEEE PES Boston

For any questions regarding this presentation or any IEEE PES Boston Technical Meetings, please contact: PES Boston Technical Meetings Chair: Zhihao ‘Ricky’ Jiang (zhihao.jiang@eversource.com)