{"slug":"swisscom-builds-enterprise-agentic-ai-for-customer-support-and-sales-on-amazon-bedrock-agentcore","url":"https://findausecase.com/use-cases/swisscom-builds-enterprise-agentic-ai-for-customer-support-and-sales-on-amazon-bedrock-agentcore","title":"Swisscom builds enterprise agentic AI for customer support and sales on Amazon Bedrock AgentCore","description":"Swiss telecom provider Swisscom uses Amazon Bedrock AgentCore (Runtime, Identity, Memory) with the Strands Agents framework to build and scale customer-facing support and sales agents. Agents are shared across departments via MCP servers and the A2A protocol, and run under Swiss data protection requirements. The blog post co-written with Swisscom staff describes the architecture and challenges rather than quantified results.","company":"Swisscom","industry":"Telecommunications","country":"Switzerland","aiCapabilities":["Agentic AI","Conversational AI"],"technology":["Amazon Bedrock AgentCore","Strands Agents"],"deployment":"Unknown","problemStatement":"Scaling AI agents enterprise-wide means managing siloed agentic solutions while coordinating across departments, with secure authentication and authorisation, interoperability of shared MCP servers and agents, and customer insight, evaluation and observability, all under Switzerland's strict data protection laws.","solutionApproach":"Swisscom used Amazon Bedrock AgentCore (Runtime, Identity and Memory) with the Strands Agents framework. It built two B2C agents, one generating personalised sales pitches and one providing automated technical-issue support, which are being integrated into its generative AI chatbot SAM and shared across the organisation through MCP and A2A.","businessValue":"According to the source, development teams reached their first business stakeholder demos within 3-4 weeks despite no prior Strands Agents experience, and one project team migrated from LangGraph to Strands Agents citing reduced complexity and faster development. The source describes the architecture and challenges rather than quantified business results.","evidence":{"band":"high"},"sourceUrl":"https://aws.amazon.com/blogs/machine-learning/how-swisscom-builds-enterprise-agentic-ai-for-customer-support-and-sales-using-amazon-bedrock-agentcore","dates":{"publishedAt":"2026-10-14T05:45:51.752Z","publishedAtSource":"pipeline","updatedAt":"2026-10-14T05:45:51.752Z"},"license":"Open for reading and citing with a link to https://findausecase.com/use-cases/swisscom-builds-enterprise-agentic-ai-for-customer-support-and-sales-on-amazon-bedrock-agentcore. Bulk republication requires permission."}