cubinCheese

All Resume Projects Dump:

🎛️ Distributed Systems / Networks Section

Distributed Database System

Examined research reports FlexiRaft and QuORAM to better understand the current ecosystem and challenges of Distributed Systems
Drafting and ideation of Architecture Design for our distributed system, based on research reports and academic teachings
Collaborated with peers and consulted professors to strengthen understanding of Distributed Systems and Network Communication Concepts
Completed and led project proposal, coordinating meetings, drafting deadlines, and further planning Project Deliverables

Multi-Threaded HTTP Webserver | Python, Threading, Socket, Logging, Json, OOP

Designed and implemented webserver support for HTTP/1.0 and HTTP/1.1 protocols with Persistent (Keep-Alive) Connections
Enabled serving of Static Files from a configurable Document Root, supporting standard GET and HEAD methods
Implemented comprehensive HTTP Status Response Handling (400, 403, 404, 405, 500) and basic Blacklist Access Control (403)
Integrated custom Logging for Connection Tracking and Server Activity
Simulated Production Traffic by cloning the SJSU Homepage using wget for testing file serving capabilities

Leader Election Project | Python, Socket, Json, UUID, OOP

Designed and implemented Leader Election Algorithm for basic Unidirectional Node Topology, and later a Double-Partial Ring Topology
Debugged and developed knowledge on Timing Challenges, Socket Programming, Multi-Threaded Programming, & Event-Based Architecture
Learned a lot about the value of having Full-Coverage Testing Environments, and Logging

Microservices with Docker | Python, FastAPI, Docker, GCloud, OOP

Developed Microservices Architecture-Based Application utilizing Docker and Google Cloud for Deployment & Hosting
Created an Aggregator to coordinate microservices for user-based GET Requests, for Mock-Database Data on grocery store items
Ensuring implementation of HTTPs Routing Structure and REST API Standard, such that academic peers can mix and exchange microservices and aggregators to test functionality.

🌐 Full-Stack / Web Section

Vivid Notes AI | Next.js, OpenAI API, Supabase, Tailwind CSS

React-Based Modern Web App built on Next.js framework that leverages OpenAI API to act as Note-Taking Assistant in real time
Implemented Login System and Data Storage utilizing Supabase (PostgreSQL + Auth), securing user-specific data
Took advantage of Git/GitHub for Version Control, Docker to manage development and deployment environments
Accelerated development and deployment using Next.js App Router, ESLint, Turbopack, PNPM, and Vercel

RainSpotter | React, Modern Web Dev

Integrated RESTful APIs into a React Web App to dynamically display Live Weather and Time Data, focusing on Content Freshness and UX
Designed and coding a Responsive UI for mobile and desktop users, smoothening performance across >95% of tested screen sizes
Resolved API Reliability Issues by diversifying Data Sources (e.g., Database, Cache Layers, Fallback APIs), boosting Data Uptime by ~50%
Utilized Git Version Control, documented issues and resolutions within GitHub Issues, and deployed through GitHub and Vercel

Mega Mini Games| Flask, JavaScript, OOP, CI/CD, Render

Developed multiple Interactive Browser Games using Flask for Routing and JavaScript for Game Logic
Refactored Game Architecture using Object-Oriented Principles to improve Code Modularity and Maintainability
Implemented Automated Testing and CI/CD Pipelines to catch issues early and streamline deployments

🤖 [ AI Section ]

Novel Analysis Recommender | Python, Pandas, NumPy, Matplotlib

Developed a Recommender System utilizing a dataset of 20,000 Novels scraped from NovelUpdates.com
Analyzed dataset and explored Correlations Between Features such as Tags, Readers, and Novel Popularity
Implemented a Similarity Model to recommend novels based on a User Selected Novel
Created Visualizations to highlight Exploratory Analysis of dataset, and sample of model outputs

2048 Game AI Agent | Python, Statistical Learning

Built an AI Agent for the game 2048, utilizing Adversarial Search, achieving scores consistently 2-3x higher than average human players
Implemented and compared multiple Search Algorithms and Data Structures (e.g. Minimax, Expectimax) to improve Move Selection, Long-Term Strategy
Employed Deep Copying to minimize Memory Overhead for simulating thousands of Game States during Recursive Tree Exploration
Optimized AI's Decision-Making by fine-tuning Game Tree Depths, balancing Computational Efficiency and Decision Quality, resulting in improvement of Decision Speed by 20% without sacrificing Score Potential

Beer Style Prediction with Machine Learning | Python, Pandas, NumPy, Sklearn, NLTK, Matplotlib

Analyzed the RateBeer Dataset (~2.8M reviews, 40k users, 110k beers) to Classify Beer Styles from User Reviews and Ratings
Engineered Features from Quantitative Ratings (ABV, appearance, aroma, taste, palate) and Textual Data (Bag-of-Words, TF-IDF)
Implemented and evaluated multiple ML Models; achieved ~89% accuracy with Bernoulli Naive Bayes, outperforming Logistic Regression, Random Forests, and TF-IDF approaches

Deep Dream Filter | Python

Implemented an Image Processing Application using TensorFlow and Google’s Inception Model to generate Image Patterns and Textures
Developed features such as Image Loading, Layer Selection, and Parameter Adjustments to impact Pattern Generation
Designed comprehensive UI Components with PyQt5, such that users can effectively change Deep Dream Parameters in real-time

📱 [ Mobile App Section ]

Mobile Zoo Navigation App | Java/Kotlin

Developed Android Mobile Navigations App for San Diego Zoo visitors to plan visits of over 400 Exhibits
Developed rich UI Features to initiate Exhibit Selections, Scene-to-Scene Transitions, and Replanning Exhibit Choices
Implemented Data Parsing and Dijkstra’s Algorithm to calculate Optimal Routes between exhibits
Developed Real-Time Software to detect and replan Off-Route Visitors using Google Location API
Utilized Unit, Integration, End-to-End Tests to resolve Recurring Bugs and Unexpected User-Behavior

Provide feedback