AWS re:Invent 2025 - Amazon's journey deploying Quick Suite across thousands of users (BIZ203)

Deploying Quick Suite Across Thousands of Users at Amazon

Overview of Quick Suite

• Quick Suite is Amazon's Agentic AI platform that enables employees to work alongside AI agents to quickly answer questions and take actions
• Key capabilities of Quick Suite include:
  • Chat agents for natural language interaction
  • Quick Research for in-depth analysis across data sources
  • Quick Site for structured data and BI
  • Quick Flows for lightweight automation
  • Spaces for team-specific knowledge and agent curation

Challenges of Scaling AI at Amazon

• As a large, complex organization with millions of employees, Amazon faced several challenges in rolling out Quick Suite:
  • Achieving broad organizational adoption rather than targeting specific teams
  • Navigating security, legal, and compliance requirements
  • Connecting disparate data sources across the enterprise
  • Driving change management and user adoption

Amazon's Approach to Deploying Quick Suite

1. Data Unification:

   • Identified the 15 most valuable data sources to integrate into Quick Suite
   • Prioritized based on employee usage and pain points

2. Security and Compliance:

   • Underwent extensive security reviews, penetration testing, and approval processes
   • Worked with global work councils to address privacy concerns

3. Scalability and Adoption:

   • Took a data-driven, persona-based approach to training and communications
   • Automatically deployed Quick Suite extensions and add-ins
   • Conducted live training sessions, created FAQs, and established feedback loops

Quick Suite Adoption and Impact at Amazon

• Over 40,000 Amazonians have created custom chat agents in Quick Suite
• Hundreds of thousands of users have:
  • Conducted over 50,000 Quick Research queries
  • Executed over 300,000 Quick Flows
• Specific examples of impact:
  • Enterprise Engineering team saw a 90% reduction in time to create sales "plays"
  • Supply Chain Finance team expanded internationally and increased writing velocity
  • Worldwide Specialist team automated meeting preparation, saving over 10 hours per week

Lessons Learned

1. Invest in data quality and preparation - an agent is only as good as the underlying data
2. Understand user intent, not just their questions
3. Focus on precision over volume of information
4. Implement robust validation frameworks to ensure accurate and reliable responses

Conclusion

Quick Suite has enabled broad adoption of Agentic AI across Amazon, driving significant productivity gains and insights for employees. The journey to scale Quick Suite required careful planning around data, security, compliance, and change management - providing valuable lessons for other enterprises looking to deploy similar AI-powered platforms.