BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.devconf.info//devconf-us-2026//speaker//EXULTM
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-2026-GCB8CP@pretalx.devconf.info
DTSTART;TZID=EST:20260924T155000
DTEND;TZID=EST:20260924T160500
DESCRIPTION:Over the summer I learned what's actually happening behind that
 : how transformers process language through word embeddings\, how temperat
 ure and probability shape outputs\, how context windows fill and why respo
 nses vary between runs. I explored the full inference stack vLLM\, LLM-d\,
  model servers and learned how MaaS layers token metering and rate limitin
 g on top to deliver AI as a managed service. I built AI skills that automa
 te real workflows and learned that the model itself is only part of the eq
 uation the agents\, skills\, and prompting around it are what make it usef
 ul. I also studied where AI breaks down concept drift\, inconsistent outpu
 ts across runs\, token limitations which made you not just a better user o
 f AI but a more critical one. AI is everywhere and most people use it with
 out thinking about it. Now when  you interact with any AI tool\, you under
 stand the mechanics behind it. Everywhere you hear about how "ai is going 
 to take over" however that is not  the case\, how in some ways it still ne
 eds humans and I learned that during my internship at Red Hat. I am open t
 o speaking about what I learned as Red Hat AI intern about generative AI a
 nd how it actually does impact a workplace environment\, and how the world
  does still need software engineers.
DTSTAMP:20260823T232121Z
LOCATION:106 (Capacity 45)
SUMMARY:AI Still Needs Us: Lessons from Inside an AI Infrastructure Team - 
 Lina Chishti
URL:https://pretalx.devconf.info/devconf-us-2026/talk/GCB8CP/
END:VEVENT
END:VCALENDAR