BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.pydata.org//pydataglobal2025//talk//NSWVT3
BEGIN:VEVENT
SUMMARY:When the Meter Maxes Out: Chernobyl Disaster Lessons for ML System
 s in Production - Idan Richman Goshen
DTSTART:20251211T130000Z
DTEND:20251211T133000Z
DTSTAMP:20260807T120008Z
UID:pretalx-pydataglobal2025-NSWVT3@cfp.pydata.org
DESCRIPTION:At 1:23 a.m. on 26 April 1986\, the RBMK-4 graphite-moderated 
 reactor at Chernobyl exploded. Every dosimeter still working inside flat-l
 ined at 3.6 R/h\, its maximum reading\, while lethal radiation raged unsee
 n. That single detail from Chernobyl is the perfect allegory for what can 
 go wrong in modern machine-learning pipelines: clipped features\, hidden d
 istribution shifts\, missing logs\, runaway feedback loops\, and more. Thi
 s talk unpacks key incidents from the disaster and map each one to an equi
 valent failure mode in production ML\, showing how silent risk creeps into
  data systems and how to engineer for resilience. Attendees will leave wit
 h a practical set of questions to ask\, signals to track\, and cultural ha
 bits that keep models (and the businesses that rely on them) well clear of
  their own meltdowns. No nuclear physics required.
LOCATION:General Track
URL:https://cfp.pydata.org/pydataglobal2025/talk/NSWVT3/
END:VEVENT
END:VCALENDAR