IBM: Development of AI software for businesses with multi-agent capabilities and specialized modernization workflows

As companies now use AI to write vast amounts of code, their software development challenges have shifted to other areas of the process: 85% of DevSecOps experts surveyed agree that AI has shifted the bottleneck from writing code to testing and validation [1]. IBM Bob is designed to deploy AI capabilities wherever software development takes place. Instead of limiting AI to a single development interface for isolated tasks, Bob provides a unified foundation on which teams can collaborate throughout the entire software development lifecycle.

For example, developers at Jack Henry, a leading provider of financial services and banking technology, faced the challenge of maintaining and further developing an extensive RPG codebase as the application portfolio grew in scope and complexity. “With IBM Bob,” explained Kevin Sligar, Chief Technical Architect at Jack Henry, “our developers can accelerate RPG development processes, improve code quality, and gain deeper insights into the system knowledge accumulated over decades, while simultaneously increasing efficiency in enhancement efforts.”

Many enterprise engineers manually select models, trying to balance cost and performance, yet still end up with inconsistent results and unpredictable expenses. Bob can now optimize the entire execution system, not just model selection. Bob assigns models to tasks, coordinates AI execution across various agents, and—through the newly introduced Bobalytics—provides enterprises with insights into productivity, quality, performance, and costs to help them optimize AI at scale.

“Bob is the platform enterprise customers have been asking for,” said Neel Sundaresan, GM, Automation and AI, IBM. “The bar for enterprise AI is no longer set at a better coding assistant. It’s about an end-to-end, agent-based development partner that works within every system development teams already use, while providing the governance, security, and cost controls that enterprises require. We developed Bob to solve the problems enterprises actually face, and the updates we’re announcing today lay the foundation for everything that’s yet to come.”

Development teams also face unique challenges when they move beyond code generation and apply AI to larger, more complex tasks, such as updating legacy applications or modernizing IBM Z, IBM i, and Java environments.

Blue Pearl, a provider of cloud solutions and consulting services, has successfully deployed IBM Bob for these types of complex projects. “We integrated IBM Bob into a program to modernize legacy systems. A project that was originally estimated to take nine months and require 14 engineers was completed in just three days,” said Saireshan Govender, Group CEO of Blue Pearl. “The most impressive result wasn’t the speed—it was the combination of operational efficiency, cost optimization, and practical results that we could rely on and build upon.”

The results of AI systems can vary depending on how they operate, which can cause significant problems in such high-risk, multi-phase projects. Structured, repeatable workflows help reduce this variability, enabling teams to deliver reliable, verifiable results at an enterprise scale.

IBM Bob now offers prebuilt workflows that teams can customize and extend to fit their own environments, ensuring that results are consistent and verifiable—regardless of who executes them. The IBM Bob Premium packages for IBM Z, IBM i, and Java modernization each feature specifically designed workflows based on IBM’s decades of expertise, optimizing AI for enterprise teams undertaking large-scale modernization projects.

What’s new in IBM Bob:

• Integrated usage visibility and cost optimization: Users can now access “Bobalytics,” a new feature that allows them to monitor usage, allocate resources, and maintain visibility to scale AI in accordance with their internal guidelines.
• Parallel, model-native tool invocation: Bob now enables models to request multiple tools in a single pass and execute them concurrently.
• Subagents manage context at scale: Every exploratory step taken by an AI—whether reading files, performing searches, or tracing functions—can overload the context window and drive up costs. Now, Bob subagents handle complex tasks in an isolated context to deliver fast responses while keeping costs under control.

The latest version of IBM Bob is available for download at bob.ibm.com/download. For more information on the new features and capabilities, visit: https://bob.ibm.com/blog/bob-v2-release-announcement.

Now Available: IBM Bob Premium Packages

For decades, IBM has been at the forefront of modernizing enterprise IT—from mainframes and IBM i systems to the Java codebases on which global enterprises rely. The first three Bob Premium Packages translate this institutional knowledge from IBM into AI-native workflows that are structured, repeatable, auditable, and specifically designed for environments that other tools are not equipped to handle.

The Premium Packages now available include:

• IBM Z: Mainframe environments form the backbone of the global banking, insurance, and retail sectors and have historically been considered particularly challenging for AI deployment. Bob now addresses this by delivering AI-native application modernization for IBM Z for the first time—including the modernization of COBOL and PL/I applications as well as JCL analysis. For more information on the Premium Package for IBM Z, visit: https://www.ibm.com/new/announcements/announcing-the-ibm-bob-premium-package-for-z 
• IBM i: For decades, IBM i has supported business-critical processes in companies worldwide. With Bob, AI-native development is now making its debut in these environments—featuring integration of remote file systems, IBM i-specific modes and tools, and workflows tailored to the operational patterns of IBM i environments. For more information on the Premium Package for IBM i, visit: https://www.ibm.com/new/announcements/introducing-the-ibm-bob-premium-package-for-i 
• Java Modernization: Enterprise Java portfolios remain among the largest and most complex modernization challenges in today’s software landscape. Bob offers AI-powered workflows for Java modernization, including migration to Java 25, large-scale refactorings, and comprehensive dependency analyses—all in a structured and repeatable manner. For more information on the Premium Package for Java Modernization, visit: https://www.ibm.com/new/announcements/announcing-ibm-bob-premium-package-for-java-modernization 

[1] GitLab. (2026). The 2026 AI Accountability Report. 

About IBM

IBM is a leading provider of global hybrid cloud and AI solutions, as well as consulting services. We help clients in more than 175 countries leverage insights from their data, optimize business processes, reduce costs, and gain a competitive edge. Thousands of governments and companies in critical infrastructure sectors such as financial services, telecommunications, and healthcare rely on IBM’s hybrid cloud platform and Red Hat OpenShift to drive their digital transformation quickly, efficiently, and securely. IBM’s groundbreaking innovations in AI, quantum computing, industry-specific cloud solutions, and consulting provide our clients with open and flexible options.

– – – – –

Related Links

👉 www.ibm.com 

Photo: IBM