BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2026//speaker//DNYXSR
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-2026-3XDMXS@pretalx.com
DTSTART;TZID=CET:20260415T101500
DTEND;TZID=CET:20260415T104500
DESCRIPTION:Time series foundation models (TSFMs) such as Chronos\, Lag-Lla
 ma\, TimesFM\, and Siemens’ own GTT have shown strong generalization cap
 abilities across diverse forecasting tasks. However\, integrating these mo
 dels into a large organization is primarily a software engineering and MLO
 ps challenge rather than a modeling one.\n\nIn this talk\, we present a re
 al-world case study based on Siemens KPI Forecast\, a Python-based forecas
 ting platform that operationalizes multiple TSFMs as reusable\, production
 -grade services. The platform integrates both open research models and Sie
 mens-developed models behind a unified API\, supporting zero-shot inferenc
 e\, fine-tuning jobs\, and fine-tuned inference depending on user needs an
 d operational constraints.\n\nWe focus on how Python is used to compose he
 terogeneous components including open and closed-source models\, internal 
 data products\, APIs\, and orchestration layers into a consistent time ser
 ies specialist user experience. The session also covers challenges operati
 ng such services within a B2B environment\, including issues related to mo
 nitoring\, versioning\, and governance.\n\nAttendees will gain practical i
 nsights into turning TSFMs into reliable Python services that scale across
  teams and use cases.
DTSTAMP:20260412T142007Z
LOCATION:Europium [3rd Floor]
SUMMARY:From Research Models to SLAs: Operationalizing TSFMs with Python - 
 Jeyashree Krishnan\, Catarina Filipe
URL:https://pretalx.com/pyconde-pydata-2026/talk/3XDMXS/
END:VEVENT
END:VCALENDAR