Join the Goergen Institute for Data Science and Artificial Intelligence (GIDS-AI) on Zoom for Building and Evaluating Agentic AI Pipelines, a virtual lunch and learn on AI with Priyanshu Rawat '25 MS, Machine Learning Engineer at the Center for Integrated Research Computing at University of Rochester.

Description: Despite growing excitement around agentic AI, the vast majority of projects never make it past the prototype stage. The leading cause isn't bad models, regulation, or lack of talent; it's that most teams never define what success looks like before they start building.

Drawing on MIT's 2025 State of AI in Business report and hands-on experience designing and deploying production AI evaluation systems, this talk makes the case that evaluation is not a final step but the foundation on which every successful agentic AI project is built. Using a real-world agentic AI use case, we will walk through how to take a concept from prototype to production and how to build feedback loops that let your system improve over time.

Attendees will leave understanding not just how to build agentic AI, but also how to measure it honestly, and what separates a working system from one that looks good in a demo but fails in the real world.

Bio: Priyanshu Rawat '25 MS is a Machine Learning Engineer at the Center for Integrated Research Computing at University of Rochester, where he builds and deploys production-grade AI systems, including agentic RAG pipelines, fine-tuned LLMs, and end-to-end ML infrastructure. He holds a master’s in data science from the University of Rochester and a Bachelor's in Computer Science.

Priyanshu has been working with large language models for over three years, with a focus on agentic systems in production environments. Prior to his current role, he worked as a Data Science intern at FLX AI, where he applied agentic AI to problems in the financial domain.