BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260202T201806Z
LOCATION:261
DTSTART;TZID=America/Chicago:20251116T092000
DTEND;TZID=America/Chicago:20251116T093500
UID:submissions.supercomputing.org_SC25_sess220_ws_eduhpcp115@linklings.co
 m
SUMMARY:An Interactive Agentic HPC Tutor for Lesson Planning, Teaching, an
 d Assessment
DESCRIPTION:Erik Pautsch (Loyola University Chicago, Argonne National Labo
 ratory (ANL)); Mengjiao Han, Joseph A. Insley, Janet Knowles, Victor Matee
 vitsi, and Silvio Rizzi (Argonne National Laboratory (ANL)); and George K.
  Thiruvathukal (Loyola University Chicago, Argonne National Laboratory (AN
 L))\n\nHigh-performance computing (HPC) education is at an inflection poin
 t, driven by agentic systems and “prompt-engineering” as a form of program
 ming. We describe an interactive tutor built from autonomous LLM-based age
 nts, each with a narrow role: planning lessons, explaining concepts, scaff
 olding code, and executing runs. Using open-source toolkits and locally ho
 sted models on leadership-class supercomputers, the tutor lets educators g
 enerate and refine parallel-programming examples in real time without exte
 rnal APIs or subscription fees. Complex workflows are composed through str
 uctured prompts rather than traditional source code, while per-agent histo
 ry summarization prevents context-window overflow and enables self-correct
 ing code generation. Requiring no proprietary services, the platform is im
 mediately deployable in institutional HPC environments and scales from sin
 gle-user sessions to classroom labs. Beyond a teaching aid, it illustrates
  how prompt-driven, multi-agent software can deliver dynamic, personalized
 , and extensible learning experiences across technical domains.\n\nRecordi
 ng: Livestreamed, Recorded\n\nRegistration Category: Technical Program Reg
  Pass, Workshop Reg Pass\n\nSession Chairs: Erik Saule (University of Nort
 h Carolina at Charlotte), Sushil K. Prasad (University of Texas at San Ant
 onio), David P. Bunde (Knox College), and Suzanne J. Matthews (United Stat
 es Military Academy)\n\n
END:VEVENT
END:VCALENDAR
