BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260624T171345Z
LOCATION:Bldg. 8 - B 102
DTSTART;TZID=Europe/Stockholm:20260701T140000
DTEND;TZID=Europe/Stockholm:20260701T160000
UID:submissions.pasc-conference.org_PASC26_sess165@linklings.com
SUMMARY:MS5H - Co-Design of Workflow Orchestration Systems for AI/ML-Enhan
 ced Scientific Workloads on HPC Infrastructure
DESCRIPTION:Organizer(s): Ozgur Ozan Kilic (Brookhaven National Laboratory
 ), and Charles Leggett (Lawrence Berkeley National Laboratory)\n\nScientif
 ic workflows increasingly combine traditional simulations with AI/ML metho
 ds, creating complex pipelines that must operate across heterogeneous, dis
 tributed infrastructure. This minisymposium examines how co-design approac
 hes—bringing together domain scientists, HPC architects, and AI researcher
 s—can address the resulting orchestration challenges. We explore three int
 erconnected themes. First, we examine intelligent workflow generation, inc
 luding how agentic AI and large language models can assist scientists in d
 eveloping and optimizing workflows for heterogeneous architectures. Second
 , we address federated infrastructure services, discussing architectural p
 atterns for unified APIs that enable workflows to seamlessly traverse inst
 itutional boundaries while respecting security and data governance constra
 ints. Third, we consider AI-ready data and model services, exploring how f
 ederated catalogs, automated curation, and model registries can accelerate
  discovery. While drawing examples from High Energy Physics—where petabyte
 -scale data and international collaborations have driven workflow innovati
 on—the patterns we discuss apply broadly to climate science, materials res
 earch, genomics, and other data-intensive domains. Our concluding panel ex
 amines future directions for co-designed scientific computing systems.\n\n
 Lessons in Scale: Exploring the Potential of AI Agents for Global Compute 
 Orchestration\n\nLarge-scale High Energy Physics (HEP) experiments, such a
 s ATLAS at the LHC, have spent decades mastering the art of the "heterogen
 eous ensemble"—knitting together geographically distributed grid sites, Ku
 bernetes clusters, HPC centers, and commercial clouds. While these systems
  have successfu...\n\n\nKaushik De (University of Texas at Arlington)\n---
 ------------------\nCo-Designed AI Services for HEP Software Understanding
  and Portability\n\nAI-ready scientific workflows require more than data p
 ipelines and model endpoints; they also require software stacks that domai
 n scientists can understand, maintain, and adapt for heterogeneous HPC sys
 tems. This talk presents CelloAI, a locally hosted AI assistant for High E
 nergy Physics software d...\n\n\nMohammad Atif (Brookhaven National Labora
 tory)\n---------------------\nFlowgentic: An Execution Coordination Layer 
 for Adaptive, Agent-Driven HPC Workflows\n\nModern scientific workflows in
 creasingly combine large-scale HPC simulations with machine learning infer
 ence, adaptive decision-making, and dynamic control flow. While agentic or
 chestration frameworks like LangGraph provide expressive abstractions for 
 goal-directed coordination, they assume low-late...\n\n\nMatteo Turilli (I
 E University)\n---------------------\nFuture Directions for Co-Designed Sc
 ientific Computing\n\nThis panel brings together perspectives from across 
 the minisymposium to examine how co-design methodologies can evolve alongs
 ide scientific requirements and emerging computing paradigms, including AI
 -driven workflow orchestration and heterogeneous HPC infrastructures. Pane
 lists will discuss governa...\n\n\nOzgur Ozan Kilic (Brookhaven National L
 aboratory)\n\nDomain: Engineering, Physics, Computational Methods and Appl
 ied Mathematics\n\nSession Chair: Ozgur Ozan Kilic (Brookhaven National La
 boratory)
END:VEVENT
END:VCALENDAR