Pega used its annual PegaWorld conference this year not just to evangelize the future of agentic AI but to show it working. If 2024 was about unveiling the vision, 2025 was about operationalizing it at scale – across platforms, partners, and real-world enterprise systems.
Held at the MGM Grand in Las Vegas, PegaWorld 2025 brought together technologists, enterprise leaders, system integrators, and customers from around the world to witness how the company is turning abstract AI ambition into structured, auditable, customer-ready outcomes. What emerged was a surprisingly coherent story – one where generative AI isn’t the star of the show, but rather the invisible scaffolding beneath faster, more responsive business systems.
The major announcements – Blueprint enhancements, partner-led customizations, orchestration layers, developer tools, and ecosystem accolades – each served a single purpose: to remove the distance between enterprise complexity and enterprise clarity.
Blueprint: The AI Engine Moves Upstream and Downstream
The heartbeat of Pega’s transformation strategy remains Pega Blueprint, the agentic workflow designer first introduced last year. But where Blueprint once focused mostly on greenfield application ideation, its capabilities have now expanded in two directions: deeper into legacy modernization and further downstream into development and deployment.
This year, Blueprint debuted the ability to ingest a much broader range of legacy artifacts: source code, outdated documentation, video recordings of user sessions, screenshots of legacy UIs, even outputs from systems like AWS Q Developer and Google’s MAT. The goal isn’t just to reverse-engineer old systems – it’s to create modern, cloud-native blueprints that skip the months-long discovery phase entirely.
In essence, Blueprint is positioning itself as the anti-technical-debt tool. That claim is supported by Pega’s own research, which found that 68% of IT leaders say legacy systems block AI adoption, 88% are concerned technical debt is slowing competitive pace, and 57% believe it causes customer defection.
This upstream power is now mirrored downstream. Updates to Pega Infinity’s App Studio showed how developers can go from Blueprint to production-grade application faster than ever, with AI copilots assisting across UX configuration, integration mapping, robotic process automation, testing, and even CI/CD alignment.
The result is something rare in enterprise AI: traceability. From idea to workflow to deployment, Blueprint is no longer just an innovation surface. It’s a delivery pipeline.
Partner Enablement Gets Codified – Literally
One of the smarter moves this year was the debut of Powered by Pega Blueprint – a meta-layer that allows partners to embed their own knowledge, methodologies, and proprietary IP directly into Blueprint itself. It’s Pega’s way of turning its partner ecosystem into a kind of distributed product team.
Accenture, EY, TCS, Cognizant, Virtusa, and others are already live with their own customized versions of Blueprint, each tuned to their sector specialties and client frameworks. These aren’t white-label portals. They are industry-specific reasoning engines that allow partners to show – not just tell – how their approaches work, using Pega’s AI scaffolding as the substrate.
This structure has the potential to turn Blueprint into a global enterprise method marketplace – one where clients aren’t choosing between generic templates, but between proven, codified expertise from the world’s biggest systems thinkers.
As TCS’s Vikram Karakoti put it, the new model enables partners to “reimagine journeys, simplify operations, and accelerate time-to-value.” That phrase – time-to-value – echoed across nearly every keynote, panel, and demo at the event.
The Fabric That Ties It All Together: Agentic Process Orchestration
The most ambitious announcement was Pega Agentic Process Fabric, a new orchestration layer that connects agents, apps, data, and users into an enterprise-wide, open agentic network.
Where many companies are busy building armies of generative AI copilots, Pega’s proposition is different: don’t build more agents – connect them. With Agentic Process Fabric, Pega shifts the narrative from agent creation to agent cooperation.
The Fabric includes:
- A discoverable agent library across enterprise systems
- Dynamic task orchestration, choosing the right agent for each job
- Support for emerging standards like Model Context Protocol and Agent-to-Agent interoperability
- Built-in governance and audit trails, enabling agents to act but never improvise beyond their lane
Most critically, the Fabric separates AI’s creative role at design time (via Blueprint) from its executional role at runtime – what Pega calls Predictable AI. This is an architecture designed for real enterprises, not demo floors.
The message was loud and clear: autonomous enterprise doesn’t mean uncontrolled enterprise.
Customer Proof: Awards That Anchor the Ambition
PegaWorld 2025 could have remained an echo chamber of platform announcements and partner demos. Instead, the Partner and Industry Awards helped ground the vision in real-world impact.
Among the highlights:
- Cognizant was named Partner of the Year for its leadership in end-to-end transformations
- TCS was recognized both for innovation and its work modernizing telecom infrastructure
- Merkle and Maantic were honored for success with Customer Decision Hub and Pega Customer Service
- LTIMindtree was celebrated for workflow automation that improved CX and operational throughput
In the Industry Awards, transformation stories came from brands like Citigroup, Vodafone, Blue Shield of California, MetLife, and Banco BV. These were not speculative pilots. They were scaled deployments solving problems in regulated, complex environments.
The takeaway: Pega’s tech is not just for the AI-curious. It’s being used by the AI-serious.
Developer Experience: AI That Teaches While It Builds
Developer enablement was another key storyline. Updates to App Studio showcased an evolved agentic layer where AI acts as mentor, translator, and co-builder.
New features included:
- An embedded AI developer agent that offers contextual guidance and real-time coaching
- Automated integration mapping for legacy and modern systems
- AI-driven RPA for non-API environments, analyzing UI behavior and auto-generating automation
- Built-in testing and UX configuration, handled by AI based on Blueprint inputs
The message here was accessibility. By fusing AI with a low-code environment, Pega is making enterprise-grade app delivery available not just to IT teams, but to operations, compliance, and customer leaders as well.
Technical Debt: From Liability to Opportunity
PegaWorld didn’t shy away from the hard stuff. One of the more sobering insights came from a joint research release with Savanta: nearly half of global enterprises are still running applications over 20 years old – and most can’t sunset them because they remain business critical.
But Pega’s take wasn’t fatalistic. Instead, the company is positioning these legacy assets as fuel for transformation. With Blueprint’s ingestion engine now able to handle documentation, video, screenshots, and source code, legacy becomes a strategic starting point – not a roadblock.
This reframing is powerful. Where others see technical debt as a liability, Pega sees optionality – and speed.
What PegaWorld 2025 Really Told Us
The defining insight of PegaWorld 2025 is this: agentic AI is only as valuable as the system that governs it.
From Blueprint to App Studio to Agentic Process Fabric, Pega is building not just intelligent agents, but the connective tissue, compliance layers, and orchestration engines that allow AI to actually work inside the messy, regulated, high-stakes world of the enterprise.
This isn’t OpenAI bolted onto ERP. It’s something else. A system-native, workflow-first, compliance-aware AI framework that brings structure to what is otherwise just suggestion.
For the thousands gathered in Las Vegas, the message wasn’t just heard. It was demonstrated.
