BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.adass.org//adass2026//speaker//YE8J83
BEGIN:VTIMEZONE
TZID:AWST
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1;UNTIL=20051231T160000Z
TZNAME:AWST
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
END:STANDARD
BEGIN:STANDARD
DTSTART:20070325T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:AWST
TZOFFSETFROM:+0900
TZOFFSETTO:+0800
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20071028T030000
RRULE:FREQ=YEARLY;BYDAY=4SU;BYMONTH=10;UNTIL=20081025T190000Z
TZNAME:AWDT
TZOFFSETFROM:+0800
TZOFFSETTO:+0900
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-adass2026-CMAGVN@pretalx.adass.org
DTSTART;TZID=AWST:20261104T152400
DTEND;TZID=AWST:20261104T153600
DESCRIPTION:In the past years the field of observatory operations has seen 
 many technical advances which have predominantly been driven by the need f
 or robotisation. Employing the new opportunities provided by AI agents adv
 ances this development. The German Centre for Astrophysics (DZA) has launc
 hed several projects in this regard. In this contribution we present\n(1) 
 Exploitation of libraries of documentation (manufacturer documentation\, t
 echnical manuals\, acceptance protocols for telescopes and receivers\, com
 missioning notations and reports) through a RAG system that incoporates a 
 knowledge graph and an agentic router and that enables queries in a chatbo
 t-like manner which bases information solely on the existing documents. Th
 is facilitates efficient access to complex technical information compiled 
 by multiple providers and teams with extended cross-referencing. An exampl
 e use case is the commissioning of the 14 element array of the MeerKAT-Ext
 ension telescopes.\n(2) Support of observatory proposal evaluation through
  a Multi-Agent Review System (MARS). It uses several specialised AI agents
  that fill different roles in a review process to create structured evalua
 tions that are designed to provide transparent pre-assessments on all aspe
 cts of a proposal. Key features include using verifiable evidence as an in
 formation base\, recording all steps that lead to a quantified assessment 
 or ranking\, and using consistent but adaptable scoring frameworks.\nWe wi
 ll present the development and current capabilities of these systems as we
 ll as the strategies used to verify the results.
DTSTAMP:20261113T210052Z
LOCATION:Seminar Room 2
SUMMARY:AI assisted observatory management - Jankowsky Felix
URL:https://pretalx.adass.org/adass2026/talk/CMAGVN/
END:VEVENT
END:VCALENDAR