Worked on the rasbt/llms-from-scratch repository to deliver a GPU-enabled AWS SageMaker Jupyter notebook provisioning feature, streamlining the setup process for large language model development. Used CloudFormation to implement Infrastructure as Code templates that automatically provision GPU-backed Jupyter environments with pre-installed LLM tooling, enabling one-click setup and reproducible experimentation. Focused on reducing onboarding and environment configuration time, the solution supports scalable machine learning workflows and consistent development practices. Demonstrated technical proficiency in AWS, Python, and YAML while integrating cloud infrastructure with data science tooling. No major bugs were reported during this period, reflecting a focused and well-executed engineering effort.