Cloud Engineer with a strong background in AWS serverless architectures, data pipelines, and infrastructure as code. I work primarily with AWS CDK to design reproducible, scalable systems, and I’m currently exploring how LLMs can be integrated into cloud platforms in a practical, engineering-focused way.
I’m not a machine learning researcher — my focus is on building reliable platforms that use AI, rather than training large models from scratch. This includes patterns such as retrieval-augmented generation (RAG), task-specific model refinement, and orchestration of LLM workflows using managed AWS services.
- Cloud & Infrastructure: AWS CDK, Lambda, ECS/Fargate, EC2, VPC, IAM
- Data Engineering: AWS Glue, Step Functions, S3, event-driven pipelines
- AI / LLM Systems (Applied): Amazon Bedrock, SageMaker (fine-tuning basics), RAG patterns
- Dev Practices: CI/CD, reproducible deployments, cost-aware design
- Designing LLM-enabled serverless architectures on AWS
- Exploring Bedrock + retrieval patterns for real-world workloads
- Building small, well-documented CDK projects that demonstrate how things fit together
Below are selected repositories that reflect how I think about cloud engineering and applied AI:
- AWS CDK infrastructure examples – VPCs, ECS, IAM, and data pipelines
- Generative AI demos – focused on integration, not hype
- End-to-end pipelines – from ingestion to processing to deployment
Each repository is intentionally scoped and documented to explain why a design choice was made.
I write about AWS and applied GenAI to clarify my own thinking and share practical lessons:
- Medium: https://medium.com/@katevu
- LinkedIn: https://www.linkedin.com/in/katevu
If you’re interested in cloud platforms, data pipelines, or practical GenAI on AWS, feel free to connect or open a discussion on any repo.
This profile reflects ongoing learning. I value clarity, correctness, and sustainable engineering over buzzwords.