Engagement Overview
Squid AI combines website analytics, visitor intelligence, and AI-assisted insights. Tactical Edge's engagement focused on connecting behavioral signals to contextual analysis and downstream sales and marketing workflows.
The Challenge
- Connect website activity and visitor signals to a useful customer workflow.
- Give users context for a signal instead of requiring them to reconcile separate analytics views.
- Integrate qualified insights with CRM workflows while retaining appropriate data-use controls.
The Approach
- Connect behavioral-data ingestion and preparation with the existing analytics product.
- Organize context for model-assisted interpretation of visitor journeys and buying signals.
- Define interfaces for downstream CRM handoff, monitoring, and operational support.
What Changed
The work connected analytics context and AI-assisted interpretation with the product's sales-intelligence workflow. Conversion impact requires a defined baseline and measurement that accounts for campaign and audience changes.
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.
An account team could ask why an account's activity changed. An agent could retrieve permitted journey events, summarize the observed behavior, and prepare a CRM follow-up task for review without treating inferred intent as an established fact.
| Service | Role in the extension |
|---|---|
| Amazon Bedrock | Summarize permitted visitor journeys and explain the observed signals behind a proposed follow-up. |
| Bedrock Knowledge Bases | Retrieve account context, product material, and permitted analytics records for a sourced response. |
| AgentCore Runtime and Gateway | Look up an account and prepare an approved CRM task using tenant-scoped tools. |
| AWS workflow services | Use Kinesis for incoming signals, Lambda for event processing, and a durable workflow for deduplication, retries, and CRM handoff. |
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
Apply consent, retention, tenant isolation, and source-system permissions before retrieval or CRM updates. Keep visitor identification and inferred intent subject to the product's approved data-use rules.
Track insight relevance, CRM handoff accuracy, duplicate tasks, reviewer acceptance, and conversion impact against a documented baseline.
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