WALTHAM, Mass. — Rocket Software today announced the expansion of Rocket EVA, its agentic AI platform for mission critical systems. The expanded platform provides governed, auditable AI agents on the mainframe to correlate and reason across operational data and automate operational tasks within enterprise-defined policies. The latest capabilities expand the operational use cases teams can address with EVA, helping enterprises put agentic AI to work across mission-critical environment
As enterprises bring AI to the mainframe, specialized expertise remains a significant challenge. According to a Hanover Research study for Rocket Software, 81% of financial services IT leaders report a very or extremely significant mainframe skills gap, while 87% believe AI will help address it over the next two years.
“Enterprises have relied on the mainframe to run their mission-critical workloads for decades,” said Milan Shetti, president and CEO of Rocket Software. “AI promises to unlock even more value from these environments, but only if enterprises can deploy it securely and without disruption. We’re helping customers apply agentic AI to mission-critical systems with speed, confidence, and control, closing the skills gap and putting that expertise within reach of every enterprise team.”
Backed by Rocket Software’s mainframe expertise, modernization portfolio and purpose-built AI architecture, EVA enables agents to understand and act on mainframe environments in real time. Global organizations across financial services, government, insurance, retail, and telecommunications are participating in EVA pilots. The focused pilot program enables customers to validate high-value operational use cases with their own data and move from installation to actionable insights within days, not weeks.
The expanded EVA platform includes the launch of Rocket PlanGuard, a new security layer that puts a policy checkpoint between AI reasoning and system execution, extending auditable, highly scoped agent access to mainframe resources. PlanGuard adds a policy decision point and identity controls to ensure AI-driven actions remain within approved boundaries while maintaining enterprise control.
Putting Agentic AI to Work Across Core Operations
According to Hanover Research, 94% of financial services IT leaders rank enhancing IT operations with AI as a high or top priority. EVA uses natural language to apply AI to operational challenges that have traditionally required specialized expertise, multiple tools and significant manual analysis, significantly lightening the burden on IT teams. According to a recent Rocket survey of mainframe AI use, Rocket customers are seeing early success with EVA across a range of high-value operational use cases, including:
- Operations/Diagnostics: Identifies and correlates operational issues across systems to accelerate analysis and resolution, with use cases spanning job failure pattern analysis, operational dashboard and KPI generation, end-to-end agentic workflow automation, end-of-month financial reporting analysis, IBM CICS online banking application optimization, and IBM CICS queue and backlog analysis.
- Security: Protects critical systems and data with vulnerability detection, automated compliance, and intelligent patch management, and identification of emerging risks associated with Post Quantum Cryptography (PQC).
- Batch Processing: Assists teams in monitoring batch performance against SLAs, identify critical failures and investigate root causes faster through AI-assisted analysis.
- Application/Configuration Management: Helps pinpoint performance and configuration issues through high CPU utilization analysis and IBM Db2 data sharing, buffer pool and cache degradation.
- Data: Advances Rocket’s vision for unified enterprise data access, enabling more intelligent delivery and streamlined output management across the enterprise.
Together, these use cases show how EVA can reduce manual investigation and accelerate the path from operational issue to root cause and resolution.
Capabilities for Mainframe Professionals
EVA combines AI-powered assistance with governed access to help mainframe professionals work more efficiently while maintaining enterprise control:
- Reduce Mean Time to Resolution (MTTR): Quickly diagnose and resolve operational issues by connecting data, context, and expertise across mainframe and distributed environments, helping teams move from problem identification to resolution faster.
- Improve Visibility: Surface insights from logs, metrics, applications, and system data through a conversational experience, making it easier to understand system health, identify dependencies, and detect issues before they escalate.
- Preserve Expertise: Capture and operationalize institutional knowledge, enabling team members of all experience levels to benefit from the expertise of seasoned professionals and reducing reliance on a shrinking pool of specialists.
- Governed Mainframe Access: Keeps AI-driven actions within patent-pending defined policy boundaries through contextual, just-in-time authorization, with activity logged for auditability and human oversight supported where required.
The expansion also extends Rocket’s model-agnostic architecture with a lightweight, standards-based approach to connecting z/OS data. By simplifying access to operational and enterprise data while maintaining governance and control, EVA enables organizations to accelerate AI adoption across core business systems with less complexity and risk.
To learn more about Rocket Software’s AI and modernization solutions, visit www.rocketsoftware.com.
SD Times Q&A
What is Rocket EVA and what does it do on the mainframe?
Rocket EVA is an agentic AI platform designed for mainframe and mission-critical systems. It uses natural language to automate operational tasks—such as job failure analysis, compliance monitoring, and batch processing diagnostics—within enterprise-defined policy boundaries. EVA correlates and reasons across operational data in real time, reducing the need for specialized manual intervention.
How does Rocket PlanGuard enforce governance for AI agents on z/OS?
Rocket PlanGuard is a security layer that sits between AI reasoning and system execution on the mainframe. It adds a policy decision point and identity controls to ensure AI-driven actions stay within approved boundaries. All activity is logged for auditability, and human oversight can be enforced where required.
How does Rocket EVA help close the mainframe skills gap?
EVA captures and operationalizes institutional knowledge so that team members at all experience levels can act on guidance previously held only by seasoned mainframe specialists. According to a Hanover Research study commissioned by Rocket Software, 87% of financial services IT leaders believe AI will help address the mainframe skills gap within two years.
Which mainframe use cases does agentic AI support in Rocket EVA?
EVA supports use cases across operations and diagnostics (e.g., job failure pattern analysis, IBM CICS optimization), security (vulnerability detection, Post Quantum Cryptography risk identification), batch processing SLA monitoring, application and configuration management (IBM Db2, high CPU analysis), and enterprise data access and output management.
How quickly can enterprises get actionable insights from Rocket EVA?
Rocket Software’s focused pilot program is designed to move customers from installation to actionable insights within days rather than weeks. Participants validate high-value operational use cases using their own production data in their own environments.
