BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2025//speaker//VNRJDV
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-TRUUVL@pretalx.com
DTSTART;TZID=CET:20250424T101500
DTEND;TZID=CET:20250424T104500
DESCRIPTION:Building and deploying scalable\, reproducible machine learning
  pipelines can be challenging\, especially when working with orchestration
  tools like Slurm or Kubernetes. In this talk\, we demonstrate how to crea
 te an end-to-end ML pipeline for anomaly detection in International Space 
 Station (ISS) telemetry data using only Python code.\n\nWe show how Kubefl
 ow Pipelines\, MLFlow\, and other open-source tools enable the seamless or
 chestration of critical steps: distributed preprocessing with Dask\, hyper
 parameter optimization with Katib\, distributed training with PyTorch Oper
 ator\, experiment tracking and monitoring with MLFlow\, and scalable model
  serving with KServe. All these steps are integrated into a holistic Kubef
 low pipeline.\n\nBy leveraging Kubeflow's Python SDK\, we simplify the com
 plexities of Kubernetes configurations while achieving scalable\, maintain
 able\, and reproducible pipelines. This session provides practical insight
 s\, real-world challenges\, and best practices\, demonstrating how Python-
 first workflows empower data scientists to focus on machine learning devel
 opment rather than infrastructure.
DTSTAMP:20260413T210259Z
LOCATION:Hassium
SUMMARY:Scaling Python: An End-to-End ML Pipeline for ISS Anomaly Detection
  with Kubeflow - Christian Geier\, Henrik Sebastian Steude
URL:https://pretalx.com/pyconde-pydata-2025/talk/TRUUVL/
END:VEVENT
END:VCALENDAR