Frequently Asked Questions (FAQs)

1. How should we manage repositories for Chatbot?

• Organize repositories based on modules:

  • Chatbot Backend

  • NLP Engine (Gen AI)

  • Terraform scripts for Deployments

  • Knowledge Content

2. How should we scale our infrastructure?

• Evaluate scaling strategies based on demand and growth patterns.

• Consider horizontal scaling for adding more nodes or vertical scaling for enhancing existing nodes.

3. Should we use Network Load Balancer or Application Load Balancer?

• Decide based on specific requirements:

  • Network Load Balancer (NLB) for even traffic distribution.

  • Application Load Balancer (ALB) for advanced routing, SSL termination, and content-based routing.

4. How should we scale Vector Database Infrastructure?

• Assess growth patterns and scaling needs.

• Implement horizontal scaling by adding more nodes or vertical scaling by enhancing existing nodes.

5. Should we use a common GenAI module for all Chatbots?

• Evaluate the benefits, considering performance, customization, and maintenance implications.

6. How to manage infrastructure?

• Implement version control for infrastructure-as-code scripts.

• Ensure proper documentation for onboarding and troubleshooting.

7. How to monitor the health of infrastructure?

• Use tools like AWS CloudWatch to monitor health.

• Set up alarms for critical metrics, e.g., EC2 instances utilization.

8. How to monitor our APIs?

• Employ API monitoring tools to track response times, error rates, and overall health.

9. How to identify performance bottlenecks?

• Conduct regular performance testing.

• Utilize profiling tools and metrics to pinpoint areas of improvement.