BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//good-2026//talk//DDCDPV
BEGIN:VTIMEZONE
TZID:MST
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10;UNTIL=20061029T090000Z
TZNAME:MST
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
END:STANDARD
BEGIN:STANDARD
DTSTART:20071104T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:MST
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000402T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4;UNTIL=20060402T100000Z
TZNAME:MDT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20070311T030000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:MDT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-good-2026-DDCDPV@pretalx.com
DTSTART;TZID=MST:20260311T164000
DTEND;TZID=MST:20260311T165000
DESCRIPTION:Large language models are now a common tool for writing code an
 d exploring ideas\, but using them on HPC systems can still be a challenge
 . To make this simpler\, I built an interactive Open OnDemand application 
 that lets users access a local LLM directly inside Jupyter Notebook using 
 the Jupyter AI extension. Behind the scenes\, user requests are routed thr
 ough a load balancer to a pool of vLLM inference servers running on Intel 
 GPUs\, which serve open-source models up to 70B parameters.\n\nIn this tal
 k\, I’ll walk through how the system works\, why we built it\, and how i
 t gives researchers easy\, reliable access to LLM assistance without relyi
 ng on external cloud services.
DTSTAMP:20260619T091815Z
LOCATION:Main Hall
SUMMARY:Accessible LLM Inference via Open OnDemand - Keegan [Texas A&M Univ
 ersity]
URL:https://pretalx.com/good-2026/talk/DDCDPV/
END:VEVENT
END:VCALENDAR