Academy Tutorial IPDPS 2026

8:30 am - 12:00pm, May 26, 2026, New Orleans, LA

Alok Kamatar and Kyle Chard

Introduction

Agentic systems, in which autonomous agents collaborate to solve complex problems, are emerging as a transformative methodology in AI. However, adapting agentic architectures to scientific cyberinfrastructure --- spanning HPC systems, experimental facilities, and federated data repositories --- introduces new technical challenges. Join us at IPDPS 2026 for a half-day tutorial, where we will introduce participants to the design, deployment, and management of scalable agentic systems for scientific discovery. We will present Academy, a Python-based middleware platform built to support agentic workflows across heterogeneous research environments. Participants will learn core agentic system concepts, including asynchronous execution models, stateful agent orchestration, and dynamic resource management. A guided hands-on session will help attendees build and launch their own agentic workflows. We will present case studies in materials discovery, biology, and chemistry. This tutorial is designed for researchers, developers, and cyberinfrastructure professionals interested in advancing AI-driven science with next-generation autonomous systems. To attend to the tutorial, you need to register to attend IPDPS.

Tutorial Materials

Code

The code for the tutorial is available in the github repository. Clone the repository and checkout the "ipdps-26" branch for material specific to this tutorial. (Note, the branch will be available closer to the tutorial date).

Slides

We are still finishing the slides, check back closer to the date of the tutorial.

Participant Requirements

Participants need intermediate knowledge of coding in python, and a laptop where they can write and run python code. A test linux environment for remote computing and access to an inference service will be provided to attendees of the tutorial. (If completing the tutorial independently, instructions for setting up the test environment will be provided, and a different inference service can be used.)

Schedule (Tentative)