BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:EDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T010000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:EST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260621T180637Z
UID:79AA3D70-B8FB-4B2B-AC93-9A8D96B82058
DTSTART;TZID=America/New_York:20260724T170000
DTEND;TZID=America/New_York:20260724T180000
DESCRIPTION:This session explores how to make autonomous AI agents trustwor
 thy at scale.\n\nUsing a fraud investigation scenario built on Apache Kafk
 a\, the Model Context Protocol (MCP)\, and an LLM-based reasoning agent\, 
 the talk shows how governance can be engineered into both the real-time da
 ta layer and the tool boundary that controls agent actions.\n\nAttendees w
 ill learn\n\n- Why the trust boundary in an agentic system belongs in the 
 server\, not only in the prompt\n- How MCP can be used as a typed\, audita
 ble governance contract\n- Engineering patterns behind shadow mode\, appro
 val-gated tools\, and multi-model routing\n- How these concepts come toget
 her in a working open-source reference implementation they can explore aft
 erward\n\nThis session is designed for engineers\, architects\, product le
 aders\, researchers\, students\, and anyone interested in building trustwo
 rthy AI-agent systems.\n\nCo-sponsored by: Central Indiana IEEE Computer S
 ociety\n\nVirtual: https://events.vtools.ieee.org/m/563554
LOCATION:Virtual: https://events.vtools.ieee.org/m/563554
ORGANIZER:siva.bayyavarapu@ieee.org
SEQUENCE:58
SUMMARY:Feeding and Fencing AI Agents: Engineering Governed Agentic Systems
  on Real-Time Data
URL;VALUE=URI:https://events.vtools.ieee.org/m/563554
X-ALT-DESC:Description: <br /><p>This session explores how to make autonomo
 us AI agents trustworthy at scale.</p>\n<p>Using a fraud investigation sce
 nario built on Apache Kafka\, the Model Context Protocol (MCP)\, and an LL
 M-based reasoning agent\, the talk shows how governance can be engineered 
 into both the real-time data layer and the tool boundary that controls age
 nt actions.</p>\n<p>Attendees will learn</p>\n<ul>\n<li>Why the trust boun
 dary in an agentic system belongs in the server\, not only in the prompt</
 li>\n<li>How MCP can be used as a typed\, auditable governance contract</l
 i>\n<li>Engineering patterns behind shadow mode\, approval-gated tools\, a
 nd multi-model routing</li>\n<li>How these concepts come together in a wor
 king open-source reference implementation they can explore afterward</li>\
 n</ul>\n<p>This session is designed for engineers\, architects\, product l
 eaders\, researchers\, students\, and anyone interested in building trustw
 orthy AI-agent systems.</p>
END:VEVENT
END:VCALENDAR