BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.devconf.info//devconf-cz-2026//speaker//DZNQYL
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-devconf-cz-2026-3ULJZZ@pretalx.devconf.info
DTSTART;TZID=CET:20260619T133500
DTEND;TZID=CET:20260619T135000
DESCRIPTION:Data teams want fast\, safe ways to ship models to production 
 — but cloud budgets and managed platforms are not always an option. This
  talk presents a simple\, fully on-prem MLOps reference setup built from o
 pen-source components: training with notebooks/Airflow\, experiment tracki
 ng and registry with MLflow + Postgres\, artifact storage on MinIO (S3)\, 
 promotion via MLflow aliases (Test/Production)\, and per-alias serving usi
 ng MLflow Serve containers.\nThe key idea: aliases act as stable contracts
 \, so promotion and rollback become an instant alias switch while model ve
 rsions change underneath. You’ll see a short demo of the workflow: train
  → register → switch alias → a new serving container starts → dash
 boards update.\nWe’ll also cover what this setup monitors today (health 
 checks\, logs via Loki\, basic Grafana panels) and what to add next (laten
 cy/RPS/error metrics\, drift and quality monitoring).
DTSTAMP:20260430T115420Z
LOCATION:A113 (capacity 64)
SUMMARY:Local MLOps on a Budget: MLflow Aliases\, Auto-Serving & Observabil
 ity - Aleksei Turov
URL:https://pretalx.devconf.info/devconf-cz-2026/talk/3ULJZZ/
END:VEVENT
END:VCALENDAR