BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pydata-amsterdam2026//speaker//TFK7JU
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-pydata-amsterdam2026-EXBMZD@pretalx.com
DTSTART;TZID=CET:20260910T111500
DTEND;TZID=CET:20260910T114500
DESCRIPTION:New items in a marketplace have no behavioral signals. Ranking 
 models ignore them. They never get booked. They never get signals. The loo
 p never breaks.\n\nThis talk is a three-year\, experiment-driven journey t
 o break that loop: what we tried\, what failed\, and what ultimately worke
 d. You'll leave with a reusable blueprint for solving cold start in any ra
 nking or recommendation system.
DTSTAMP:20260710T150504Z
LOCATION:Unconference
SUMMARY:Cold Start at Scale: Three Years of Experiments in a Travel Marketp
 lace - Theodore Meynard
URL:https://pretalx.com/pydata-amsterdam2026/talk/EXBMZD/
END:VEVENT
END:VCALENDAR