IBM Bob bolsters, boosts & broadens AI-driven software application development
IBM called its agentic software development platform Bob.
Was this because Big Blue wanted another three-letter acronym to sit alongside IBM perhaps? Bob could mean Best of Breed. Or it could relate to Bob the Builder to tally with the “build” process that software engineers engage in.
Equally, it could just be an anthropomorphism to make Bob sound as human as Claude (which derives its name from Claude Shannon, obviously).
Actually, it’s none of those, apparently IBM likes to keep that part a secret.
What’s not a secret are Bob’s new multi-agent capabilities, built-in AI cost and use analytics tools, and its pre-built specialised workflows for modernising enterprise systems.
According to IBM, now that organisations are using AI to write massive amounts of code, their software development challenges have moved to other parts of the process with 85% of DevSecOps professionals surveyed agreeing that AI has shifted the bottleneck from writing code to reviewing and validating it.
“IBM Bob is architected to bring AI capabilities wherever software engineering work happens. Rather than limiting AI to a single development interface for isolated tasks, Bob provides a unified foundation for teams to coordinate across the software development lifecycle,” states the company.
Accelerated development workflows
Using IBM Bob, developers are said to be able to accelerate development workflows, improve code quality, and gain deeper insights into decades of accumulated system knowledge while gaining efficiency.
IBM’s Sundaresan: Bob can now optimise across the execution system, not just model selection.
Bob can now “optimise across the execution system”, not just model selection.
This notes is made in relation to the fact that the execution system is the entire operational environment where Bob orchestrates multi-agent workflows. It coordinates parallel tool calling, isolates context windows using subagents, and dynamically routes specific tasks to the most cost-effective models.
Matches models a la mode
Bob matches models to tasks, coordinates AI execution across agents, and provides organisations with visibility into productivity, quality, performance, and cost through the newly launched Bobalytics, to help enterprises optimise 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 a better coding assistant. It’s an end-to-end agentic development partner that works inside any system development teams already use, with the governance, security, and cost controls enterprises require. We built Bob to solve the problems enterprises actually have, and the updates we’re announcing today are the foundation for everything that comes next.”
Engineering teams also encounter unique challenges as they move beyond code generation and apply AI to larger, more complex work like updating legacy applications or modernising IBM Z, IBM i, and Java environments.
High-stakes, multi-phase projects
AI output can vary depending on how the work is done, which can create significant issues for high-stakes, multi-phase projects.
Structured, repeatable workflows help reduce that variability so teams can deliver reliable, auditable results at enterprise scale.
IBM Bob now has pre-built workflows available that teams can customise and extend for their own environments to ensure outcomes are consistent and auditable, regardless of who runs it.
Further, IBM Bob Premium Packages for IBM Z, IBM i, and Java Modernisation are each opinionated workflows built on decades of IBM’s domain experience that optimise AI for enterprise teams that need to do large-scale modernisation.
