
The Journey to Kiu: From an Idea to a Global Solution
Have you ever wanted to build a virtual assistant but felt overwhelmed by AI and serverless concepts?
- Deliver real-time information about AWS User Groups and their events.
- Facilitate connections within the tech community.
- Demonstrate that with the right tools, there are no limits to what we can achieve.
- Where will the event be held?
- How can I become a sponsor?
- How do I become a speaker?
- Essential concepts:
- Providing information about communities and events, using only four services: Amazon API Gateway, Amazon Bedrock, AWS Lambda, and DynamoDB.
- Automated community registration: Previously, adding a community required manual intervention. Now, with an API Gateway, AWS Lambda, and DynamoDB-based registration system, communities can self-manage within the platform.
- Expansion of communication channels: Initially, Kiu was only integrated with WhatsApp, but to improve accessibility, it now supports web platforms using React and Amplify.
- Support for multiple languages: With the help of advanced Prompt Engineering techniques and Amazon Bedrock's natural language processing capabilities, Kiu can now answer questions in different languages.
- Cost Optimization: Thanks to adopting an event-based model with Amazon EventBridge, we reduced unnecessary Lambda invocations, minimizing operational costs.
- Integration with Meetup and Sessionize: This allows organizers to automatically import events and improve talk and speaker planning. Alexandra Fernández wrote an article about her experience creating the integration with Sessionize, which you can read here: https://community.aws/content/2sYaliat3cCNISUcV5R78pB9Frq/kiu-y-sessionize-transforming-event-management-in-aws-user-groups
- Use of Cognito for authentication: This facilitated access management and permissions within the Kiu platform.
- How to connect a Bedrock Agent with WhatsApp.
- The challenge of integrating Sessionize and Meetup into an AI agent.
- Prompt Engineering and multi-language support.
- Performance and scalability.
- Observability and monitoring.
- Cost and optimization.