Programmed a developer tool for engineers to visualize the interaction of microservices in Kubernetes clusters of Docker containers. Errors and broken communication between microservices are denoted in red, helping developers to debug A/B tests.
(Learn more...)Applied machine learning to categorize user-uploaded images and renamed the files accordingly, allowing users to download a ZIP file of the renamed images.
(Learn more...)Created a trivia game with thousands of random questions pulled from an API for users who have registered an account and answered demographic questions. After each game, users can see relational graphs of scores v.s. user demographics, including age, education level, and location. I also architected a live chat for clients to communicate with others connected to the server.
(Learn more...)Mood Buddy is a free iOS app I've created and published on the App Store because I care deeply about people's mental health. Want to listen to unguided meditation, log your workouts, write journals, and share your mood score with your Mood Buddy? Self care is best when it's a daily habit. Download Mood Buddy to take care of yourself in the midst of COVID19.
DownloadUsed agile methodology and a scrum board to construct an app that solves the issue of drivers wasting time looking for parking. Parked drivers can indicate where they parked and when they're leaving, while drivers looking for a spot can reserve the spot in advance. Both the giver and the receiver will see each other's car information for an easier transfer of the parking spot.
(Learn more...)Engineered a shooting game with strong OOP concepts, class inheritance, and polymorphism. Players use the arrow keys to move and space bar to shoot different types of spaceships that fire different weapons. Health points change upon consuming a first-aid kit or hit by an enemy. All entities appear at random positions from the right side of the screen, where different spaceships move to the left at different speeds. Each level gets harder.
(Learn more...)
Applied Fast Fourier Transformation (FFT) on audio data to train a supervised neural network with 99.98% accuracy of identifying monophthong voice inputs.
Used C language to import neural net configuration, implement matrix calculation (w = ax + b
), and instruct an Arduino
to lock or unlock my handmade box.
This product encompasses computer science, electrical engineering, and mechanical engineering skills.
A Discord AI bot that flags hateful messages using Natural Language Processing (NLP) and Naive Bayes algorithm. It would respond to Islamophobic hate speech with educational texts from the Quran. The website is hosted as a GitHub page as a standalone React.js application with visualization of flagged messages using Chart.js.
(Learn more...)