BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//txlf2026//speaker//GQJBDV
BEGIN:VTIMEZONE
TZID:US/Central
BEGIN:STANDARD
DTSTART:20251106T000000
TZNAME:CST
TZOFFSETFROM:-0600
TZOFFSETTO:-0600
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260308T030000
RDATE:20270314T030000
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T020000
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:RamaLama + Podman Quadlets — Build Task-Specific AI Services wit
 h systemd - David Duncan
DTSTART;TZID=US/Central:20261106T133000
DTEND;TZID=US/Central:20261106T145000
DTSTAMP:20261119T212852Z
UID:pretalx-txlf2026-88EDHE@pretalx.com
DESCRIPTION:Running a model locally is easy. Turning that model into multi
 ple task-specific services — each with its own security boundary\, syste
 m prompt\, and restart policy — is where most guides stop. This workshop
  picks up where `ramalama run` leaves off.\n\nRamaLama gives you container
 -native local AI inference with one command: pull a model from an OCI regi
 stry\, detect your GPU\, serve it. Podman quadlets turn that into a system
 d-managed service that starts on boot\, restarts on failure\, and logs to 
 journald\, but one model can do many jobs! By placing lightweight agent co
 ntainers in front of a single RamaLama model server you will build task-sp
 ecific AI services that behave like any other daemon on your Linux box. No
  Kubernetes required.\n\nPrerequisites: Laptop with a browser. Familiarity
  with Linux command line helpful but not required — we explain every uni
 t file directive.
LOCATION:Stadium
URL:https://pretalx.com/txlf2026/talk/88EDHE/
END:VEVENT
END:VCALENDAR
