Fanatics Betting and Gaming (FBG) developed a multi-agent customer support system on AWS to address the unique challenges of sports betting. Customers demand instant, accurate answers, especially during live events when every minute counts. The system handles complex queries about account issues, deposit limits, state-specific regulations, and responsible gaming resources, which vary across jurisdictions. Traditional chatbots struggle with this complexity, often frustrating customers and increasing costs as human agent queues grow. FBG’s engineering team built a multi-agent AI system on AWS that resolves customer issues faster, more accurately, and at a fraction of the cost of human-only support. Source: awsml
FBG, a sports betting platform part of the Fanatics family of brands, operates across multiple U.S. states and serves a rapidly growing user base that demands around-the-clock support, especially during high-traffic events like the NFL playoffs and the Super Bowl. The team faced exponential growth in support volume, prompting the need for a more efficient system. The challenge included handling diverse queries from transaction history to betting rules and self-exclusion options, with no single model or knowledge base covering it all. Operators must also identify and respond to signs of problem gambling in real time, requiring nuanced understanding of conversational context beyond keyword matching. FBG needed a system that could handle this complexity autonomously while knowing when to escalate to a human agent. Source: awsml
Rather than relying on a single monolithic chatbot, FBG designed a multi-agent system where specialized agents handle different aspects of customer interactions. The team leveraged their existing Amazon Elastic Kubernetes Service (Amazon EKS) platform to deploy, scale, and iterate on each agent independently. They chose Amazon Bedrock for its model-agnostic access to multiple foundation models through a single API, allowing them to match tasks to the best model and swap models as better options emerge. The system also inherits FBG’s established security and governance controls, with Amazon Bedrock Guardrails providing responsible AI safeguards for compliance requirements. Source: awsml
