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:20260605T005352Z UID:1294A6C4-4410-46BF-96E7-126DF68AB337 DTSTART;TZID=America/New_York:20260817T190000 DTEND;TZID=America/New_York:20260817T200000 DESCRIPTION:As enterprises increasingly adopt generative AI to improve prod uctivity\, customer experience\, software development\, knowledge manageme nt\, and decision-making\, the efficient use of tokens has become a critic al factor in controlling cost\, improving performance\, and scaling AI sys tems responsibly. This talk explores practical strategies for building tok en-efficient enterprise AI solutions without compromising accuracy\, secur ity\, or user experience.\n\nThe session will discuss how token consumptio n impacts latency\, infrastructure cost\, model performance\, and scalabil ity in real-world enterprise environments. It will cover techniques such a s prompt optimization\, context reduction\, retrieval-augmented generation \, model selection\, response compression\, caching\, batching\, and intel ligent routing across multiple AI models. The talk will also highlight gov ernance considerations\, including monitoring token usage\, measuring retu rn on investment\, enforcing data privacy\, and aligning AI deployments wi th business objectives.\n\nAttendees will gain a clear understanding of ho w enterprises can move from experimental AI pilots to production-ready\, c ost-aware\, and scalable generative AI systems. The session is especially relevant for technology leaders\, engineers\, architects\, and AI practiti oners focused on maximizing the value of AI while maintaining operational efficiency and responsible AI practices.\n\nSpeaker(s): Ravi\, \n\nVirtual : https://events.vtools.ieee.org/m/561661 LOCATION:Virtual: https://events.vtools.ieee.org/m/561661 ORGANIZER:anushanerella@ieee.org SEQUENCE:53 SUMMARY:WIE Speaker Series: Enhancing Real-Time Data Reliability through Ob servability-Driven Streaming Architecture in Distributed Systems URL;VALUE=URI:https://events.vtools.ieee.org/m/561661 X-ALT-DESC:Description: <br /><p>As enterprises increasingly adopt generati ve AI to improve productivity\, customer experience\, software development \, knowledge management\, and decision-making\, the efficient use of token s has become a critical factor in controlling cost\, improving performance \, and scaling AI systems responsibly. This talk explores practical strate gies for building token-efficient enterprise AI solutions without compromi sing accuracy\, security\, or user experience.</p>\n<p>The session will di scuss how token consumption impacts latency\, infrastructure cost\, model performance\, and scalability in real-world enterprise environments. It wi ll cover techniques such as prompt optimization\, context reduction\, retr ieval-augmented generation\, model selection\, response compression\, cach ing\, batching\, and intelligent routing across multiple AI models. The ta lk will also highlight governance considerations\, including monitoring to ken usage\, measuring return on investment\, enforcing data privacy\, and aligning AI deployments with business objectives.</p>\n<p>Attendees will g ain a clear understanding of how enterprises can move from experimental AI pilots to production-ready\, cost-aware\, and scalable generative AI syst ems. The session is especially relevant for technology leaders\, engineers \, architects\, and AI practitioners focused on maximizing the value of AI while maintaining operational efficiency and responsible AI practices.</p > END:VEVENT END:VCALENDAR