BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.devconf.info//devconf-us-2025//talk//WAJTLA
BEGIN:VTIMEZONE
TZID:EST
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10;UNTIL=20061029T070000Z
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:STANDARD
DTSTART:20071104T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000402T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4;UNTIL=20060402T080000Z
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20070311T030000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-devconf-us-2025-WAJTLA@pretalx.devconf.info
DTSTART;TZID=EST:20250920T125000
DTEND;TZID=EST:20250920T130000
DESCRIPTION:Choosing the right technology stack is critical as the prevalen
 ce of AI-based products grows. Countless options exist for models\, runtim
 es\, operating systems\, hardware\, and more. Identifying which pieces wor
 k together best for a particular use case is crucial to ensure optimal per
 formance.\n\nThe project lets users specify and simulate their target AI s
 tack. They can then compare different configurations side-by-side–for ex
 ample\, two different models–to determine which variation is more suitab
 le. These results are then visualized and aggregated into an interpretable
  format\, enabling the users to make informed development decisions.\n\nIn
  this talk\, you’ll learn about how models can be deployed on the edge\,
  what metrics are relevant\, and how AI models differ in size and performa
 nce. You’ll also learn how to run models that you can interact with righ
 t on your local machine.
DTSTAMP:20260311T004929Z
LOCATION:106 (Capacity 45)
SUMMARY:Characterizing AI inferencing stacks - Ryan Malone
URL:https://pretalx.devconf.info/devconf-us-2025/talk/WAJTLA/
END:VEVENT
END:VCALENDAR