Sr. Applied AI Solutions Architect - Public Sector, Amazon Connect

Job Overview

This position is part of the AWS Specialist and Partner Organization (ASP). Specialists own the end-to-end go-to-market strategy for their respective technology domains, providing the business and technical expertise to help our customers succeed. Partner teams own the strategy, recruiting, development, and growth of our key technology and consulting partners. Together they provide our customers with the expertise and scale needed to build innovative solutions for their most complex challenges.

The Applied AI Solutions Architecture team within AWS is seeking a hands‑on, customer‑obsessed Solutions Architect to accelerate customer adoption of Amazon Connect's AI capabilities.

• Customer Engagement:
  Lead technical discovery sessions with customer teams to understand business requirements, existing contact center architecture, and AI readiness. Translate findings into actionable implementation plans.
• Customer Data Readiness:
  Conduct data readiness assessments to evaluate the quality, accessibility, structure, and governance of customer data assets (CRMs, knowledge bases, ticketing systems, order management, etc.). Identify data gaps, recommend remediation strategies, and help customers build the data foundation required for effective AI agent tool use and RAG‑powered responses.
• Agentic AI Implementation:
  Design and configure agentic AI solutions within Amazon Connect, including AI agent creation, AI prompt engineering, model selection, guardrail configuration, and tool/action integration.
• MCP Server Configuration:
  Design and deploy Model Context Protocol (MCP) servers that expose customer tools, data sources, and APIs in a standardized format — enabling AI agents to dynamically discover and invoke capabilities across the customer's technology stack.
• A2A (Agent‑to‑Agent) Integration:
  Architect Agent‑to‑Agent communication patterns that allow Amazon Connect AI agents to collaborate with specialized agents across the enterprise (e.g., billing, logistics, IT) enabling multi‑agent workflows that span organizational boundaries.
• Integration Development:
  Build serverless integrations using AWS Lambda, API Gateway, Step Functions, and scripting (Python, Node.js) to connect Amazon Connect AI agents with customer data systems (CRMs, ERPs, databases, knowledge bases).
• Cloud Data Access:
  Architect secure access patterns to cloud-based data systems (Amazon Dynamo
  
  DB, Amazon RDS, Amazon S3, Amazon Open Search, Amazon Kendra/Knowledge Bases for Bedrock) to power AI agent tool use and retrieval‑augmented generation (RAG).
• Pre‑Production Validation:
  Guide customers through testing, evaluation, and validation of AI agent performance against defined success criteria before production deployment.
• Knowledge Sharing:
  Create reusable artifacts (reference architectures, implementation guides, sample code, prompt libraries, data readiness checklists) that scale best practices across the Connect SA community and partner ecosystem.
• Service Team
  
  Collaboration:
  
  Provide feedback to Amazon Connect and Amazon Bedrock product teams based on real‑world customer implementations, contributing to product roadmap prioritization.

• Pair‑programming with customer developers to build and test AI agent configurations.
• Designing prompt strategies and evaluating model performance across different foundation models.
• Configuring MCP servers to expose customer APIs, databases, and tools in a standardized format for agent consumption.
• Designing A2A workflows where Amazon Connect agents hand off to or collaborate with specialized agents across the customer's enterprise.
• Configuring knowledge bases and data connectors for RAG‑powered agent responses.
• Conducting architecture reviews and providing prescriptive guidance for production readiness.
• Documenting implementation patterns and contributing to the team's knowledge base.

• 6+ years of design, implementation, or consulting in applications and infrastructures.
• Experience with Amazon Connect or other enterprise contact center platforms (Genesys, Avaya, Cisco, NICE, Five9, etc.).
• Hands‑on experience with Amazon Bedrock, including model invocation, agent…