Agentic AI Frontier Seminar
Frontiers of Agentic AI — research, systems, and practice
Online · Pacific Time (PT)

Agentic AI Frontier Seminar

A seminar series on Agentic AI: models, tools, memory, multi-agent systems, online learning, and safety, featuring leading researchers and industry experts.

Join us! Register here to get seminar updates.

Recordings (with speaker consent) will be posted on our YouTube channel.

View Upcoming About the Seminar

Incoming Seminar

Online · 2026-02-27 · 09:00–10:00 PT

Talk Title: Rethinking Post Training

Associate Professor · Pulkit Agrawal · MIT

Talk Abstract: In this talk, I will discuss the limitations of current post-training methods and discuss recent advances in continual learning that holistically study: (i) What data should the agent learn from? (ii) What is the right learning algorithm? (iii) and finally questioning the pre-training / finetuning paradigm. This talk will span our papers RL Razor, Self-Distillation Learning, SEAL, and more recent work.

Bio: Pulkit Agrawal is an Associate Professor in the Department of Electrical Engineering and Computer Science at MIT. He earned his Ph.D. from UC Berkeley and co-founded SafelyYou Inc. Pulkit completed his bachelor’s from IIT Kanpur and was awarded the Director’s Gold Medal. His work has received multiple Best Paper Awards, the IEEE Early Career Award in Robotics and Automation, the IROS Toshio Fukuda Young Professional Award, the IIT Kanpur Young Alumnus Award, the Sony Faculty Research Award, the Salesforce Research Award, the Amazon Research Award, the Signatures Fellow Award, the Fulbright Science and Technology Award, and others.

Join Link More details

Focus Areas

Foundation Models & Core Capabilities

Agent Infrastructure & Tooling

Learning, Adaptation & Feedback

Multi-Agent Systems & Social Intelligence

Evaluation, Safety & Alignment

Applications & Vertical Use Cases

Interface & Interaction Design

Governance, Ethics & Ecosystem Building

Organizing Committee

Photo of Ming Jin

Ming Jin

Virginia Tech

He is an assistant professor in the Bradley Department of Electrical and Computer Engineering at Virginia Tech. He works on trustworthy AI, safe reinforcement learning, foundation models, with applications for cybersecurity, power systems, recommender systems, and CPS.

Email Website
Photo of Shangding Gu

Shangding Gu

UC Berkeley

He is a postdoctoral researcher in the Department of EECS at UC Berkeley. He works on AI safety, reinforcement learning, and robot learning.

Email Website
Photo of Yali Du

Yali Du

KCL

She is an associate professor in AI at King’s College London. She works on reinforcement learning and multi-agent cooperation, with topics such as generalization, zero-shot coordination, evaluation of human and AI players, and social agency (e.g., human-involved learning, safety, and ethics).

Email Website