US: +1 (610) 247-9595 / EU: +48 666 74 71 74
US: +1 (610) 247-9595 / EU: +48 666 74 71 74
ul. Antoniego Pajdaka 5/98
03-134 Warsaw, Poland
NIP : 8921405210
Regon : 146293701
Created by Quality Pixels | Privacy Policy
Engineering AI HubIBM Engineering AI Hub is the AI platform for IBM Engineering Lifecycle Management (ELM). It brings AI into the tools where engineering work already happens, with the controls you need to use it across teams and programmes.
Specialist AI agents help engineers write better requirements, catch up on long work items, draft test cases and start system models. Each agent works with live project data in IBM Engineering Requirements Management DOORS Next, IBM Engineering Workflow Management, IBM Engineering Test Management and IBM Engineering Rhapsody, not with copies pasted into a chat window.
Governed MCP tools let your own AI assistants and IDEs search, analyse and update ELM data under the same role-based permissions your teams use today.
Engineers review what AI produces before it changes an artifact, and every AI interaction is recorded. You decide where it runs, in the cloud, on premises or air-gapped, and which model provider it uses.
Administrators choose the agents, MCP tools and endpoints on offer. Usage reports show adoption by period and unique users, and rate limits keep use fair as it grows. Agent activity and tool calls stay on record too, and changes trace back to the artifacts they touched.
Find the line closest to your team. Each one shows what changes with Engineering AI Hub.
Capabilities
Each requirement is scored against industry standards. The agent names the problems, such as ambiguity or missing conditions, and proposes clearer wording for the author to accept or reject.
IBM Engineering Requirements Management DOORS Next
The Engineering Assistant answers questions, summarises, searches topics and translates, from your project data.
IBM Engineering Requirements Management DOORS Next
A short synopsis of a long task, defect or feature, and new work items drafted from a few words.
IBM Engineering Workflow Management
Test cases and scripts drafted from the requirements in a test plan, with pre- and post-conditions.
IBM Engineering Test Management
Use cases and actors found in requirements become model elements and diagrams, linked back.
IBM Engineering Rhapsody Systems Engineering
Governed MCP tools give IDEs and assistants access to ELM data, under existing permissions.
IBM Engineering Lifecycle Management
Ask about a specification in plain language and get answers drawn from your project data.
Context, key details, status, working notes and risks of a long-running work item.
A few words become a draft epic, feature or story with acceptance criteria.
Test cases with pre- and post-conditions and detailed scripts, new in 1.4.
Use cases and actors found in requirements, then created in the model.
Usage by period, unique users, agents and MCP tools, exported to PDF.
Engineering AI Hub is the control plane for AI in IBM ELM: four sets of controls, managed in one place.
Who can use AI?
Access managed centrally, with your existing ELM sign-in and rolesWhat data can AI see?
Only the engineering context the user is allowed to see
What can AI do?
You choose the agents, MCP tools, endpoints and models on offerWho approves a change?
An engineer reviews AI output before it is applied
How much is AI used?
Adoption and usage reports by period and unique usersWill it stay stable as use grows?
Rate limits and resource controls for fair use
What did AI do?
Interactions, agent activity and tool calls kept as recordsWhat changed because of AI?
Changes traced to the artifacts they touched
IBM Engineering Requirements Management DOORS NextQuality analysis, Engineering Assistant
IBM Engineering Workflow ManagementWork item synopsis and compose
IBM Engineering Test ManagementTest generation
Engineering AI HubAgents and MCP tools on top of your ELM data and permissions
IBM Bob, Copilot and other MCP clients work with ELM data.
watsonx Orchestrate and others call IBM agents through A2A.
IBM watsonx.ai, Amazon Bedrock or Azure AI.
Tokens shared with your other ELM applications.
Trusted in regulated industries
Our clients build products under strict standards and regular audits. We supply licences, implementation, hosting and training.
Our clients build products under strict standards and regular audits. We supply licences, implementation, hosting and training.
We are a partner and reseller of these technology providers.
Services
With IBM tokens, from the same token pool as your other ELM applications. The model provider is licensed separately: watsonx.ai, for example, as a cloud subscription with usage, or on premises with your own GPU hardware. We size both from the agents and teams you plan to start with.
In the cloud, on premises or air-gapped, on Red Hat OpenShift or Kubernetes. It needs IBM ELM 7.1 or later with Jazz Authorization Server, an IBM Common Licensing server with AI Hub licences, a PostgreSQL database, Redis and a model runtime.
Agents and MCP tools use your existing ELM sign-in and role-based permissions, so AI sees only what the user may see. Requests go to the model runtime you choose, which can run in your own data centre or in an air-gapped network.
Your engineers. AI creates and people evaluate, or people create and AI evaluates. AI-generated changes can require human review before they are applied, and every interaction is recorded.
Yes. The governed MCP tools give MCP clients such as IBM Bob or Copilot access to requirements, work items, tests, models and source control. Administrators decide which tools and endpoints are switched on.
Start with one team and one agent. We set up the environment with you, connect it to your ELM sign-in and token pool, and track usage with the built-in reports before you roll it out further.
They discover, analyse, act and trace across the lifecycle. For requirements: search and browse, analyse the impact of change, assess coverage and gaps, create and link. For work items and source control: search, compare changes, track iteration status, create, comment and link. For tests: search, analyse coverage and results, assess release readiness, comment and link. For models: search, explore structures, navigate branches and versions, analyse dependencies. IBM Documentation lists all MCP tools in Engineering AI Hub 1.4.
A request starts from an ELM application, an AI assistant or IDE, or an orchestrated workflow. The Supervisor agent intercepts it and delegates the task to the appropriate endpoint. The agent or tool invokes a Large Language Model on watsonx.ai, Amazon Bedrock or Azure AI, and returns the response in its interface.
Discover: find engineering artifacts and uncover trusted context. Analyse: generate insights from engineering data and relationships. Act: create and update artifacts and links through AI-assisted actions. Automate: build AI agents and skills that orchestrate multi-step workflows.
Add them to the ELM Mini Dashboard and hand repetitive requirements work over to them. Each one previews its changes first.
Browse all widgetsLeave your details and we will get back to you, usually within one business day.
