Engagement Overview
Tyler Union manufactures waterworks products including ductile iron fittings, joint restraints, and valve boxes. The Tactical Edge engagement focused on an engineering-knowledge workflow that connects technical information with service guidance.
The Challenge
- Find relevant manuals, specifications, procedures, and service information across distributed sources.
- Help engineers and technicians verify that guidance applies to the product and revision in question.
- Retain review and traceability when model-assisted guidance enters an engineering workflow.
The Approach
- Organize approved engineering content for retrieval and connect model responses to supporting sources.
- Define tools and workflow interfaces for engineering and service tasks.
- Keep higher-impact guidance within the team's existing review and operating controls.
What Changed
The engagement connected engineering knowledge retrieval and service-guidance workflows around approved technical content. The source document, product revision, and engineering review remain necessary when applying a recommendation.
AWS Agentic AI Capability Extension
The following architecture describes capabilities that could extend this engagement's workflow. It distinguishes possible additions from the original customer work.
A technician could ask about a procedure for a particular product revision. An agent could retrieve the applicable manual, identify missing context, and draft a service note for an authorized reviewer.
| Service | Role in the extension |
|---|---|
| Amazon Bedrock | Explain retrieved engineering material and draft a response that names the applicable product context. |
| Bedrock Knowledge Bases | Retrieve approved manuals, specifications, and procedures with document and revision references. |
| AgentCore Runtime and Gateway | Read product or service records and prepare a reviewed service note through approved APIs. |
| AWS workflow services | Use S3 for approved source material, Lambda for product-system integrations, and CloudWatch to monitor retrieval and tool execution. |
AgentCore can host an application's agent and expose approved tools through Gateway, with Identity for service access and Observability for execution traces. The application defines its tool permissions, validation, and approval steps.[1]
Bedrock Knowledge Bases can retrieve relevant source material and include citations in generated responses. Select the retrieval configuration and apply the source content's access rules before supplying it to a model.[2]
Existing Amazon Bedrock Agents deployments can use action groups and knowledge bases for orchestration. AWS now calls this service Bedrock Agents Classic and has closed it to new customers; new agent implementations should evaluate AgentCore.[3]
Bedrock Guardrails offers configurable content, sensitive-information, and grounding checks. Test the selected checks against representative requests and keep authorization and human approvals in the application workflow.[4]
Operating Boundaries and Evaluation
Do not substitute generated text for an approved engineering procedure. Require review when guidance affects installation, safety, or product configuration, and preserve source-system permissions.
Evaluate retrieval relevance, document-revision accuracy, unsupported answers, tool-call success, and the time needed to produce reviewed guidance.
AWS References
These references document the extension's AWS capabilities.
- Amazon Bedrock AgentCore overview. https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html
- Amazon Bedrock Knowledge Bases. https://docs.aws.amazon.com/bedrock/latest/userguide/knowledge-base.html
- Amazon Bedrock Agents Classic. https://docs.aws.amazon.com/bedrock/latest/userguide/agents.html
- Amazon Bedrock Guardrails. https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails.html