BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//scipy-2026//talk//8QMU8G
BEGIN:VTIMEZONE
TZID:CST
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10;UNTIL=20061029T080000Z
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
BEGIN:STANDARD
DTSTART:20071104T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:CST
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000402T030000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=4;UNTIL=20060402T090000Z
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20070311T030000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:CDT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-scipy-2026-8QMU8G@pretalx.com
DTSTART;TZID=CST:20260715T131500
DTEND;TZID=CST:20260715T134500
DESCRIPTION:Field inspections in agriculture and science face a common prob
 lem: hands are full\, data needs structure\, and decisions require both in
 spection history and domain expertise. I built HiveGuide\, an open-source 
 field inspection system with three main components: (1) voice transcriptio
 n for hands-free data entry\, (2) AI extraction to structured data and act
 ion items\, and (3) an AI assistant that provides intelligent advice by ro
 uting between personal inspection history and authoritative domain literat
 ure. For the assistant\, I tested 7 routing strategies on 500+ queries to 
 solve a dual-source problem: when to query your data versus domain referen
 ces. The LLM classifier approach balanced accuracy and speed without requi
 ring training data. The architecture is transferable to any inspection dom
 ain where you need minimal device interaction and intelligent advising.
DTSTAMP:20260715T023320Z
LOCATION:Memorial Hall
SUMMARY:AI-Powered Field Inspection: Voice Capture\, Data Extraction\, and 
 Intelligent Multi-Source Routing - Carolyn Olsen
URL:https://pretalx.com/scipy-2026/talk/8QMU8G/
END:VEVENT
END:VCALENDAR