Week 4 Worklog

Week 4 Objectives:

  • The goal of this week was to strengthen understanding of AWS data and AI services — from building scalable DataLake architectures to working with serverless NoSQL databases (DynamoDB) and exploring AI development lifecycle management.
  • Additionally, this week aimed to enhance technical translation and communication skills through hands-on labs, blog translations, and participation in AWS events.

Tasks to be carried out this week:

DayTaskStart DateCompletion DateReference Material
1Building a Data Lake on AWS
- Data Lake Concepts: Explored centralized repository architectures for varied data analytics workloads.
- AWS Glue (ETL): Studied Glue Crawlers for automated schema discovery and Data Catalog generation.
- Amazon Athena: Performed serverless SQL queries directly on S3 data.
- Amazon QuickSight: Visualized data insights via interactive Dashboards.
- Hands-on Lab 35: Deployed Data Lake infrastructure, ingested data via Kinesis Firehose, and queried cataloged data using Athena.		29/09/202529/09/2025Module 07
2Query Optimization & Cost Analysis (Lab 40)
- ETL Workflow: Executed end-to-end pipeline: Raw Data Ingestion -> Glue Crawler -> Transformation (Parquet) -> Athena Query.
- Cost Analysis: Analyzed AWS billing data using SQL in Athena (Service breakdown, Tagging).
- Optimization Techniques: Applied cost-saving strategies: Parquet compression, query LIMITs, and data partitioning.		30/09/202530/09/2025Module 07
3NoSQL Databases with Amazon DynamoDB (Lab 60)
- DynamoDB Fundamentals: Researched Serverless NoSQL architecture, auto-scaling, and Capacity Modes (On-demand vs Provisioned).
- Data Modeling: Deep dived into Primary Keys (Partition + Sort) and Secondary Indexes (GSI/LSI).
- Consistency Models: Compared Eventual Consistency vs. Strong Consistency trade-offs.
- Implementation: Practiced table creation, item manipulation, and querying via AWS Console and SDK (Boto3).		01/10/202501/10/2025Module 06
4AWS Technical Blog Research & Translation
- Generative AI: Translated and studied GenAI lifecycle management using MLflow on SageMaker.
- Industry Application: Researched Deep Learning applications for subsurface infrastructure mapping.
- Contact Center AI: Explored Amazon Connect optimization using AI capabilities.
- Skill Development: Enhanced technical vocabulary and domain knowledge in SageMaker, Batch, and ParallelCluster.		02/10/202502/10/2025
5AWS Event: AI Development Lifecycle & Kiro
- AI Lifecycle: Gained insights into the full AI development pipeline: Data Prep -> Training -> Deployment -> Monitoring.
- AWS AI Ecosystem: Understood the role of SageMaker, Bedrock, and CodeWhisperer in AI DevSecOps.
- New Tooling (Kiro): Explored Kiro for unified AI workflow management.
- Real-world Use Cases: Analyzed case studies on AI automation and model optimization in production.		03/10/202503/10/2025

Week 4 Achievements:

  • Built an AWS DataLake pipeline integrating Glue, Athena, and QuickSight, gaining hands-on experience in data ingestion, transformation, and visualization.
  • Configured AWS Glue Crawlers and Athena queries for cost analysis and schema automation, applying cost optimization strategies (e.g., Parquet, partitioning, query limits).
  • Mastered DynamoDB fundamentals, including primary/composite keys, indexes (GSI, LSI), consistency models, and capacity modes.
  • Enhanced translation and comprehension skills by translating AWS technical blogs on AI, MLflow, and HPC, deepening understanding of SageMaker, Batch, ParallelCluster, and Connect.
  • Participated in an AWS event focused on the AI Development Lifecycle and Kiro, gaining insights into end-to-end model management, versioning, and deployment best practices.
  • Improved technical vocabulary and practical knowledge in cloud computing, AI model governance, and data-driven architecture design within the AWS ecosystem.