Who Am I?
Cloud Engineer with hands-on experience building AWS serverless applications, event-driven architectures, and data pipelines.
I specialize in Python automation and AWS services like Lambda, S3, SNS, EventBridge, and IAM. I focus on building scalable, cost-effective cloud solutions. Currently completing my AWS Solutions Architect certification with a strong background in serverless architecture, machine learning, and full-stack development.
Technical Skills
AWS
Lambda, S3, SNS, EventBridge, IAM, CloudWatch - building serverless and event-driven architectures
Python
Automation, scripting, data processing, and building serverless functions and full-stack applications
JavaScript
Frontend development, API integration, and interactive web applications
Development Frameworks
Streamlit, Flask, FastAPI, and Gradio for building interactive applications
Machine Learning
PyTorch, Neural Networks, Computer Vision, and OCR implementation
Linux
Ubuntu, Amazon Linux - command line operations and system administration
Architecture & APIs
Serverless, event-driven architectures, data pipelines, REST APIs, and API integration
Certifications
AWS Solutions Architect Associate
In Progress - Expected November 2025
Projects
Cloud & Infrastructure
NBA Statistics Data Lake
Built an automated pipeline using AWS Lambda and EventBridge to collect and process real-time NBA game statistics. Secured the data in S3 with IAM policies using least-privilege access. The system processes 100+ games per season with automated scheduling and data transformation, eliminating manual data collection.
Game Day Notification System
Created a serverless notification system that delivers real-time NBA game alerts via SMS and email using AWS SNS. Lambda functions trigger automatically via EventBridge to send pre-game and live game notifications for multiple teams with sub-2-second latency. Improved error handling to reduce failed notifications by 95%.
Cloud Resume Challenge
Built a serverless resume website using AWS services including S3, CloudFront, Lambda, and DynamoDB. Implemented visitor counter with API Gateway integration and automated CI/CD deployment pipeline.
AI & Language Learning
AI-Powered OCR Learning Platform
Built a Siamese neural network using PyTorch and ResNet34 to recognize Toki Pona pictographic symbols with 85% accuracy. Created an interactive web app with Streamlit that has three learning modes: OCR practice, text adventure, and sentence construction. Trained the model on Google Colab GPU and optimized it to 200ms per prediction. The system supports 11 symbol classes with 512-dimensional embeddings.
Japanese Writing Practice Platform
Built a language learning app with a Flask backend and Streamlit frontend for AI-powered writing practice. Integrated MangaOCR to recognize handwritten text and GPT to generate sentences and provide grading. The app manages practice sessions, tracks progress, and connects the frontend to backend services through RESTful API endpoints.