BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2025//talk//3CYZUH
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2025-3CYZUH@pretalx.com
DTSTART;TZID=CET:20250424T161500
DTEND;TZID=CET:20250424T164500
DESCRIPTION:Many managed MLOps platforms\, while convenient\, often fall sh
 ort in providing flexibility\, requiring complex integrations\, and causin
 g vendor lock-in. In this talk\, we’ll share our experience transitionin
 g from managed MLOps tools to a self-hosted solution built on Kubernetes. 
 We’ll focus on how we leveraged open-source tools like Feast\, MLflow\, 
 and Ray to build a more flexible\, scalable\, and customizable platform th
 at is now in use at Rewe Digital. By migrating to this self-hosted archite
 cture\, we gained greater control over our ML pipelines\, reduced our depe
 ndency on third-party services\, and created a more adaptable infrastructu
 re for our ML workloads.
DTSTAMP:20260411T140953Z
LOCATION:Europium2
SUMMARY:Building a Self-Hosted MLOps Platform with Kubernetes - Josef Nagel
 schmidt
URL:https://pretalx.com/pyconde-pydata-2025/talk/3CYZUH/
END:VEVENT
END:VCALENDAR