Worklog Week 11

Week 11 Objectives:

  • Participate in AWS Cloud Mastery Series #2 to continue resolving specialized technical issues.
  • Refactor and standardize the Frontend structure for improved stability and maintainability.
  • Implement a Multi-Stack architecture to optimize deployment speed and Serverless project management.
  • Integrate basic CRUD functionalities with AI Image Processing (using Rekognition) into the website.
  • Completely resolve deployment errors (especially CORS issues) to stabilize the system.
  • Workshop: AI Service Architecture, Security & IAM, Monitoring - Complete serverless AI pipeline setup.

Tasks to be Deployed This Week:

DayTaskStart DateCompletion DateResources
Sun- Participate in AWS Cloud Mastery Series #2 (Nov 17th): Continue receiving guidance and addressing deeper technical questions about authorization errors and the AI workflow.17/11/202417/11/2024Mentor, AWS Cloud Mastery Series
Mon- Frontend Structure Unification and Refactor: Hold team meeting to standardize the Frontend code structure for maintainability.
- Research Multi-Stack Solution: Begin analyzing how to split the template.yaml file into smaller Stacks (Multi-Stack) to optimize the sam deploy process.		18/11/202418/11/2024Serverless Architecture Docs
Tue- Implement Multi-Stack Architecture: Start splitting and configuring separate Stacks (e.g., API Backend Stack, Frontend Hosting Stack).
- Proceed with AI Image Processing Integration: Combine basic CRUD functions with image processing logic (e.g., calling Rekognition API/S3 trigger) in preparation for the Update function.
- Workshop Activity: Set up API Gateway REST endpoints and SQS queues for asynchronous AI processing.		19/11/202419/11/2024Backend Codebase, AWS Rekognition, Workshop 5.7
Wed- Error Encountered after AI Integration: The system faced errors after combining AI functionality, necessitating a full Stack deletion and redeployment.
- Leader Develops Backup Stack: The team leader created a separate, optimized Multi-Stack as a contingency and reference for future optimal deployments.		20/11/202420/11/2024Leader’s Backup Stack
Thu- Persistent CORS Error: After redeploying, the CORS issue re-emerged.
- In-depth CORS Debugging: Spent time thoroughly analyzing the root cause and permanently fixing the CORS error, ensuring correct header configuration on both API Gateway and Lambda.
- Workshop Activity: Configure IAM roles and policies for Lambda functions, set up Secrets Manager for API keys, and implement WAF rules.		21/11/202421/11/2024API Gateway/Lambda Configuration, Workshop 5.9
Fri- Team Meeting and Project Stabilization: Held a team meeting to review the new Frontend structure, stabilize the main project Stack, and synchronize the fixes for CORS and basic template errors.
- Optimization for Maintenance: Finalized the solution to use a separate stack (developed by the leader) for flexibility and easier optimization in future development.		22/11/202422/11/2024New Structure Report

Week 11 Achievements:

  • Participated in AWS Cloud Mastery Series #2, gaining deeper knowledge of Serverless, Rekognition, and solutions for authorization errors.
  • Successfully refactored the Frontend and standardized the overall project structure, improving maintainability.
  • Implemented the Multi-Stack architecture (or at least established a reliable solution/backup stack), which speeds up deployment and simplifies resource management.
  • Completely resolved the persistent CORS error after identifying the root cause, ensuring stable communication between Frontend and Backend.
  • Acquired knowledge on fixing basic Template errors and gained a clearer understanding of AWS SAM deployment issues.
  • Developed a separate stack for backup/optimization, enhancing project flexibility and safety during future major updates.
  • The project has moved into the AI functionality testing phase, and although errors were encountered, a clear path for troubleshooting has been established.

Workshop Progress - AI Services, Security & Monitoring:

  • Configured API Gateway REST API with three endpoints: /writing/evaluate, /speaking/evaluate, /flashcard/generate
  • Set up SQS queues for asynchronous message processing (writing-queue, speaking-queue, flashcard-queue)
  • Deployed Lambda functions: writing_evaluator, speaking_evaluator, rag_flashcard with proper IAM roles
  • Configured DynamoDB tables: bandup-evaluations and bandup-flashcard-sets for storing AI results
  • Integrated Amazon Bedrock (Titan Embeddings V2) and Google Gemini API for AI processing
  • Set up AWS Secrets Manager for secure API key storage
  • Configured IAM roles with least-privilege permissions for Lambda functions
  • Implemented AWS WAF rules for application-level protection
  • Set up CloudWatch Logs and Alarms for monitoring Lambda execution and errors
  • Configured CloudWatch Insights for log analysis and troubleshooting