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:20260901T045801Z UID:1496A2E1-A145-40EB-B7E3-3DBCB301F440 DTSTART;TZID=America/New_York:20260904T180000 DTEND;TZID=America/New_York:20260904T190000 DESCRIPTION:From Assistants to Agents: Building Governed AI Work on Pega\n\ nTech Talk Abstract\n\nGenerative AI in the enterprise has reached an awkw ard middle age. The demos are impressive\, the pilots are plentiful\, and yet most organizations still struggle to put an AI agent anywhere near a p rocess that actually matters — a loan decision\, a claim\, a dispute\, a servicing request. The gap is not model capability. It is trust\, traceab ility\, and cost. This session looks at how Pega approaches that gap\, and what it means for teams building on the platform today.\n\nWe will start with where generative AI already lives inside Pega: as design-time acceler ation rather than runtime guesswork. Blueprint turns a described process i nto a build-ready application design\, Infinity Studio brings AI into the development environment itself\, and Knowledge Buddy grounds answers in an organization's own content instead of an open-ended model. The through-li ne is Pega's Predictable AI architecture\, which pushes the heavy reasonin g into design time so that runtime stays token-efficient and repeatable.\n \nFrom there we move to agents. Infinity 26 lets developers connect and or chestrate both Pega and third-party agents inside governed workflows\, so agents can join mission-critical processes with control\, compliance\, and predictable outcomes. Agentic Process Fabric provides the orchestration l ayer\, with support for Model Context Protocol and Agent-to-Agent so agent s can coordinate across applications and platforms\, and with security con trols that regulate which agents talk to whom. MCP now works in both direc tions: external harnesses and assistants can invoke Pega workflows\, and P ega can reach outward. We will unpack what that architecture buys you — and what it asks of you in exchange.\n\nThe session then turns to the unco mfortable part. Agentic projects fail for mundane reasons: unbounded token spend\, inconsistent outputs\, no audit trail\, and code nobody wants to own. Pega's own research put the defect rate of AI-generated code at rough ly 1.7 times that of human-written code — tolerable in a prototype\, una cceptable in a system that pays or denies a customer. We will look at wher e a deterministic workflow should stay deterministic\, where an LLM genuin ely adds value\, and how human-in-the-loop checkpoints should be designed rather than bolted on afterward. Pricing gets the same treatment: Pega's f lat fee per resolved case rather than per token changes the economics of s caling\, and it is worth understanding why.\n\nWe will close with a worked example — an agent built to analyze business requirements and produce l evel-of-effort estimates\, grounded in a real connector catalog and real e stimation rules. It is a small\, unglamorous use case\, and that is the po int: it shows how knowledge grounding\, guardrails\, and honest scoping ma tter more than model choice\, and why the right framing for these agents i s assistance to expert practitioners rather than replacement of them.\n\nA ttendees will leave with a clear map of Pega's GenAI and agentic capabilit ies\, a set of criteria for deciding which work is a good candidate for an agent\, and a realistic view of the governance and cost questions to answ er before anything reaches production.\n\nSpeaker(s): Mr. Cherukuri\, \n\n Agenda: \n6.00 pm - Introduction\n\n6.05 pm - Presentation\n\n6.45 pm - Q& A Session\n\n6.55 pm - Vote of thanks and Conclusion\n\nVirtual: https:// events.vtools.ieee.org/m/575169 LOCATION:Virtual: https://events.vtools.ieee.org/m/575169 ORGANIZER:mguduri@ltu.edu SEQUENCE:37 SUMMARY:From Assistants to Agents: Building Governed AI Work on Pega URL;VALUE=URI:https://events.vtools.ieee.org/m/575169 X-ALT-DESC:Description: <br /><p class="MsoNormal" style="margin-bottom: 4. 0pt\; text-align: center\;" align="center"><strong><span style="font-size: 15.0pt\; line-height: 115%\;">From Assistants to Agents: Building Governe d AI Work on Pega</span></strong></p>\n<p class="MsoNormal" style="margin- bottom: 16.0pt\; text-align: center\;" align="center"><em><span style="col or: #595959\;">Tech Talk Abstract</span></em></p>\n<p class="MsoNormal" st yle="text-align: justify\;">Generative AI in the enterprise has reached an awkward middle age. The demos are impressive\, the pilots are plentiful\, and yet most organizations still struggle to put an AI agent anywhere nea r a process that actually matters &mdash\; a loan decision\, a claim\, a d ispute\, a servicing request. The gap is not model capability. It is trust \, traceability\, and cost. This session looks at how Pega approaches that gap\, and what it means for teams building on the platform today.</p>\n<p class="MsoNormal" style="text-align: justify\;">We will start with where generative AI already lives inside Pega: as design-time acceleration rathe r than runtime guesswork. Blueprint turns a described process into a build -ready application design\, Infinity Studio brings AI into the development environment itself\, and Knowledge Buddy grounds answers in an organizati on's own content instead of an open-ended model. The through-line is Pega' s Predictable AI architecture\, which pushes the heavy reasoning into desi gn time so that runtime stays token-efficient and repeatable.</p>\n<p clas s="MsoNormal" style="text-align: justify\;">From there we move to agents. Infinity 26 lets developers connect and orchestrate both Pega and third-pa rty agents inside governed workflows\, so agents can join mission-critical processes with control\, compliance\, and predictable outcomes. Agentic P rocess Fabric provides the orchestration layer\, with support for Model Co ntext Protocol and Agent-to-Agent so agents can coordinate across applicat ions and platforms\, and with security controls that regulate which agents talk to whom. MCP now works in both directions: external harnesses and as sistants can invoke Pega workflows\, and Pega can reach outward. We will u npack what that architecture buys you &mdash\; and what it asks of you in exchange.</p>\n<p class="MsoNormal" style="text-align: justify\;">The sess ion then turns to the uncomfortable part. Agentic projects fail for mundan e reasons: unbounded token spend\, inconsistent outputs\, no audit trail\, and code nobody wants to own. Pega's own research put the defect rate of AI-generated code at roughly 1.7 times that of human-written code &mdash\; tolerable in a prototype\, unacceptable in a system that pays or denies a customer. We will look at where a deterministic workflow should stay dete rministic\, where an LLM genuinely adds value\, and how human-in-the-loop checkpoints should be designed rather than bolted on afterward. Pricing ge ts the same treatment: Pega's flat fee per resolved case rather than per t oken changes the economics of scaling\, and it is worth understanding why. </p>\n<p class="MsoNormal" style="text-align: justify\;">We will close wit h a worked example &mdash\; an agent built to analyze business requirement s and produce level-of-effort estimates\, grounded in a real connector cat alog and real estimation rules. It is a small\, unglamorous use case\, and that is the point: it shows how knowledge grounding\, guardrails\, and ho nest scoping matter more than model choice\, and why the right framing for these agents is assistance to expert practitioners rather than replacemen t of them.</p>\n<p class="MsoNormal" style="text-align: justify\;">Attende es will leave with a clear map of Pega's GenAI and agentic capabilities\, a set of criteria for deciding which work is a good candidate for an agent \, and a realistic view of the governance and cost questions to answer bef ore anything reaches production.</p><br /><br />Agenda: <br /><p>6.00 pm - Introduction</p>\n<p>6.05 pm - Presentation</p>\n<p>6.45 pm - Q&amp\; A S ession</p>\n<p>6.55 pm - Vote of thanks and Conclusion</p> END:VEVENT END:VCALENDAR