Engineering Lifecycle Management has always been about connecting engineering work: requirements to tests, changes to configurations, and engineering decisions to the artifacts and processes that follow.

IBM Engineering Lifecycle Management Suite – AI Edition helps make these connections and engineering artifacts available for AI-assisted engineering. It enables a foundation for AI to work with engineering context across the lifecycle, with specialist agents, governed MCP tools, enterprise AI runtimes, and controls designed for engineering environments using ELM.

I am pleased to introduce Engineering AI Hub 1.4. In this release, the foundation expands AI-assisted engineering into verification, enhances MCP capabilities through configuration-aware tooling, increases model-provider flexibility, and gives administrators more control over AI deployment.

Let’s take a closer look.

Generate test assets directly from requirements

Requirements provide engineering intent. Turning that intent into effective tests, however, can still require significant manual effort.

The new Test generation agent helps bridge that gap. Engineers can select requirements and generate test cases using the associated lifecycle context. The generated test case includes test design, preconditions, postconditions, and detailed test scripts, while maintaining traceability to the requirements that informed them.

This creates a more connected AI-assisted workflow: requirements provide the engineering intent, AI provides a first draft of the validation approach, and engineers review and refine the result to prepare for verification activities.

Apply configuration-aware MCP tools across engineering disciplines

AI depends on access to the right engineering context. Engineering AI Hub V 1.4 expands the MCP tool catalog while improving how tools understand engineering context. New and existing MCP tools help work with global configurations, enabling them to understand project-specific configurations, variants, and process definitions. The MCP tools catalog also provides new tools for creating work items and comparing SCM artifacts.

Users can now filter application-specific MCP tools based on the task at hand. This allows AI interactions to focus on the specific engineering domain relevant to the task instead of pulling broader lifecycle context than necessary. This typically results in more relevant responses, better alignment with project configurations, reduced context noise, and more focused workflows.

Analyze quality of requirements up to 5000 characters

The Requirements quality analysis agent now supports requirements up to 5000 characters, an increase from 1500 characters to enable analysis of substantially larger requirement descriptions without breaking them into smaller sections.

This enhancement allows teams to use AI-assisted quality analysis on more lengthy and detailed requirements while maintaining existing validation controls.

Azure AI expands model runtime choice

Azure AI support is one of the significant additions in Engineering AI Hub 1.4. Organizations can now align AI Hub with their existing Azure AI investments and enterprise architecture while addressing their organizational and compliance requirements.

This expands the flexibility already available through support for IBM watsonx.ai and Amazon Bedrock. The result is a clearer separation between the AI capabilities delivered through Engineering AI Hub and the model runtime an organization chooses to support those capabilities.

New governance controls for managing deployments

As AI becomes part of engineering workflows, administration needs to extend beyond configuring individual AI capabilities.

Engineering AI Hub 1.4 gives administrators more control over the availability of capabilities of the platform for their end users by enabling or disabling MCP and A2A endpoints.

Managing LLM providers is now simpler. Administrators can configure LLM provider connections directly through the Engineering AI Hub console, reducing setup complexity and streamlining operational tasks. For teams managing evolving provider strategies, this improvement removes several steps that were required during creation of instance during install.

These controls help organizations determine how AI capabilities are exposed within their engineering environment while maintaining the governance expected for enterprise ELM deployments.

Better visibility into adoption and usage

AI adoption needs to be measurable, particularly when organizations are managing enterprise licensing and expanding usage across teams.

Engineering AI Hub 1.4 enhances usage reporting with time-based views covering today, the current and previous month, and the current and previous quarter. Reports include unique-user metrics and can be exported to PDF, providing greater visibility into adoption, engagement, utilization, and license consumption.

Scaling Engineering AI Hub for enterprise deployments

As more users and AI-assisted workflows come online, the platform supporting those experiences needs to scale with them.

Redis integration in version 1.4 enables Engineering AI Hub to operate across multiple pods with shared state management and distributed caching. This provides a stronger foundation for resilient, scalable deployments as organizations expand their use of AI across engineering.

Final thoughts

With version 1.4, Engineering AI Hub can help engineers create, analyze, and act across more of the lifecycle – from requirements and testing to engineering data and configurations, while keeping engineering judgment, governance, and control at the center.

We’re excited to see how your ELM teams use these AI capabilities to accelerate verification activities, improve traceability, and unlock greater value from AI-assisted engineering. Explore Engineering AI Hub 1.4, identify and share the ELM workflows where context-aware AI can help your teams work more effectively.

Bhawana Gupta
Senior Product Manager, IBM Engineering