IBM Bob Unveils Major Updates: Multi-Agent AI, Cost Analytics, and Enterprise Modernization Workflows

A cornerstone of the latest IBM Bob update is the introduction of advanced multi-agent capabilities. This paradigm moves beyond single-task AI assistants, enabling Bob to orchestrate and coordinate the execution of multiple AI agents and subagents. This means complex development tasks can now be broken down and handled in parallel, with each agent leveraging its own context, tools, and skills.

This multi-agent architecture allows for 'parallel, model-native tool calling,' where Bob can request and run several tools in a single turn. Furthermore, 'subagents manage context at scale,' which is crucial for efficiency. Every exploratory step an AI takes—whether it's reading files, performing searches, or tracing functions—can quickly bloat the context window and drive up operational costs. By having subagents handle intricate work in isolated contexts, Bob delivers faster responses while effectively managing costs and maintaining clean context across even the largest projects.

This agentic approach is a direct response to the growing complexity of enterprise software engineering, where meaningful tasks often require more than just a code snippet. It allows developers to delegate focused tasks to AI, freeing them to concentrate on higher-value work, ultimately accelerating the delivery of complex engineering projects.

A critical challenge for enterprises adopting AI in their development workflows is gaining clear visibility into usage, performance, and, most importantly, cost. IBM addresses this with the launch of Bobalytics, a new built-in feature providing comprehensive AI cost and use analytics.

Bobalytics offers organizations detailed insights into consumption, resource allocation, productivity, quality, performance, and overall spend. This empowers enterprises to scale their AI development work effectively while adhering to internal governance, budgeting, and compliance mandates. Developers and administrators can monitor how IBM Bob contributes to their codebase, track 'Bobcoin' spending (a conceptual representation of AI resource usage), and observe team adoption across the organization.

The platform's ability to optimize across the entire execution system, not just model selection, is a significant advantage. Instead of engineers manually balancing cost versus performance by choosing different models and often ending up with inconsistent outcomes and unpredictable spend, Bob can now intelligently match models to tasks and coordinate AI execution. This ensures that AI resources are utilized optimally, driving efficiency and predictability in AI-driven development.

Recognizing the unique and often daunting challenges of modernizing legacy enterprise systems, IBM Bob now offers pre-built, specialized workflows through new Premium Packages. These customizable workflows are designed to bring AI-native application modernization to historically complex environments like IBM Z, IBM i, and Java codebases.

• IBM Z Modernization: Mainframe environments, which are central to global banking, insurance, and commerce, have traditionally been difficult for AI to assist. Bob's Premium Package for IBM Z introduces AI-native capabilities for COBOL and PL/I modernization, alongside JCL analysis. This aims to simplify and accelerate the transformation of these mission-critical systems.
• IBM i Modernization: For the decades-old IBM i platform, Bob brings AI-native development with features like remote file system integration, IBM i-specific modes and tools, and workflows tailored to the operational patterns of IBM i shops. This supports modernization efforts for languages such as RPG, COBOL, CL, SQL, and DDS.
• Java Modernization: Enterprise Java portfolios represent some of the largest and most intricate modernization challenges. The Java Modernization Premium Package provides AI-guided workflows for seamless upgrades to Java 25, large-scale refactoring, and comprehensive dependency analysis. It helps teams coordinate upgrades, UI modernization, Liberty replatforming, and security remediation across vast Java estates.

These packages translate IBM's decades of institutional knowledge into structured, repeatable, and auditable AI-native workflows, purpose-built for environments where other tools often fall short.

IBM Bob is architected to integrate AI capabilities across the entire software development lifecycle, rather than confining AI to isolated tasks within a single development interface. This unified foundation helps teams coordinate more effectively across planning, coding, testing, security, and maintenance.

The platform embeds security into everyday development with real-time analysis and automated checks that detect vulnerabilities during code authoring, rather than post-development. For example, Bob can integrate with external security tooling like Snyk to perform SAST code analysis and SCA dependency scans, identify additional vulnerabilities, and even automatically fix critical issues like SQL injection risks, then re-scan to verify the fixes.

The ultimate goal is to boost developer productivity and accelerate modernization efforts. By automating complex, time-consuming work—from building new features to strengthening secure delivery—IBM Bob enables teams to ship high-quality software with less manual effort. Examples cited include a legacy modernization program projected to take nine months with 14 engineers being completed in three days using IBM Bob, demonstrating significant efficiency gains.