BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//juliacon-2026//talk//RSTAHL
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-juliacon-2026-RSTAHL@pretalx.com
DTSTART;TZID=CET:20260814T120000
DTEND;TZID=CET:20260814T121500
DESCRIPTION:DyadAgent is an AI coding assistant for generating and debuggin
 g Dyad code across modelling and simulation workflows\, enabling engineers
  to express complex model requirements in natural language. Evaluating the
  performance of such an agent requires verification of generated simulatio
 n results against standards of physical correctness and numerical accuracy
 . DyadAgentBench is an evaluation infrastructure designed to measure the a
 gent's modelling and simulation capabilities in a systematic and reproduci
 ble manner. In this talk\, we present the infrastructure and evaluation fr
 amework developed for DyadAgent\, covering how agent performance is assess
 ed and how the resulting insights are used to benchmark and guide iterativ
 e improvements.
DTSTAMP:20260710T093126Z
LOCATION:Tent — RW1
SUMMARY:DyadAgentBench: An Agent Evaluation Framework for Dyad Agent - Ashu
 tosh Bharambe
URL:https://pretalx.com/juliacon-2026/talk/RSTAHL/
END:VEVENT
END:VCALENDAR