Dafa M

• Developed a system capable of estimating user pose (skeleton) during an online exam session using machine learning and computer vision.
• The pose can be used to detect various gestures, such as when the test takers are frequently turning their head.
• With JavaScript, the system can run natively on browser. Thus using only the user's hardware capability for calculation (no need for centralized server for calculation).
• Low hardware requirements means the system can run well without disrupting test takers.

Skills used: Computer vision, JavaScript, Machine learning

Skills used:

• Converted a generic and cheap joystick to have a gyroscopic functionality (that's only present in more expensive controller).
• Allowing for more precise aiming even without aim-assist, similar level of granularity to that of a mouse (game development).
• Using Arduino, some sensory, and JavaScript to fulfill the project.

Skills used: JavaScript, Arduino

• The data in this project is already acquired using prior project about scraping Reddit.
• Using Matplotlib in Python to do data visualization of various data, taken from different perspectives.
• Discovered various insights from something that's seemingly random, due to the intuitive nature of the visualization.

Skills used: Python, Data visualization, Matplotlib

• Designed an app with easily scale-able functionalities and data management with the help of Google Sheets.
• App is made using Android Studio. App data is transmitted to Google Sheets through the help of Google API.
• UI Design approaches are taken here to make sure the app matches the design language set by Android Developers.

Skills used: UI Design, Google API, Android Studio

• Scarping context-rich Reddit discussion data, maintaining the conversational tree branches plus the attributes they have.
• Using Google Colaboratory + Python to host the scraping process, thus take away the workload from the client computer
• Using BeautifulSoup and Selenium as web scrapers, with Selenium more into actual user input and BeautifulSoup as a deeper dive into the web pages.

Skills used: Python, Web scrapers, Selenium

• Developed the gesture recognition subsystem using Python and Tensorflow (machine learning).
• Developed the authorization subsystems which include face recognition and clothing identification (computer vision). 
• United all the different subsystems developed by other engineers (system integration). 
• Directed other developers both in technical and documentation aspects (as a lead developer).
• Tested the hardware specifications of the system including finding the minimum requirements to run it. 

Skills used: Python, Tensorflow, Computer vision, Machine learning

• As a lab assistant, trained and helped first-year students during their lab courses on basic programming materials
• Topics: Python, Pandas, data visualization, data analytics, etc.

Skills used: Python, Pandas, Data visualization, Data Analytics

• As the head of electrical division, coordinated, designed, developed, and implemented electrical components of the robot, including basic robotics programming.
• Handling the creation of the robot's main board, and overall system integration.
• Collaborated with the research, programming, management, and mechanical divisions to further develop the robot. 
• Designed the standard procedures to handle the robot (disassembly, assembly, testing, etc). 
• Became the leader in the smaller field team of three during the day of the competition for final preparations. 

Skills used: Robotics

• Developed an oximeter with a smartphone interface (Android Studio).
• Data is processed using an analog circuit and Arduino (digital signal processing), and modulated to be able to combine 2 different signals into one.
• Data is transmitted through an analog signal to audio jack to increase the device versatility and to suppress price.
• Data is capable to be copied to be processed further (or stored) in the Android app as one of many functionalities in the app (UI Design).
• The analog circuit is simulated using SPICE (ic design) before the physical hardware is created.

Skills used: SPICE (ic design), UI Design, Arduino, Android Studio

• High-paced final project for the robotics course with various products being presented.
• Designed the product parametrically using Fusion 360 CAD (Autodesk) with acrylic as the base material. 
• Brainstormed ideas about the methods and technical specifications of the products to be designed, together with two other developers. 
• Advised and helped other developers in implementing the electrical and programming aspects of the project. 

Skills used: Autodesk

• The project is given by a senior student at the time, and taken in a team of three as a member of electrical workshop.
• The product is basically a punching bag fitted with a device capable of calculating the force applied to it using sensors, physical characteristics, and physical formulas. 
• Developed the interface using Android Studio to control, debug, and display the data of the power calculator device connected through Bluetooth. 
• Developed the communication data scheme and the dummy device for testing (iOT).

Skills used: iOT, Android Studio

• The game is a audio-based rhythm game, instead of the common visual-based. Users asked to repeat the music once they hear the example sound.
• Unity Game Design and Development.
• Implemented advertisements and analytics.

Skills used: Unity Game Design and Development

• Made a short 3D movie using Cinema4D utilizing texturing, 3D modeling, post-processing, rendering, etc.
• Learned a big deal about computer graphics, which are then utilized when developing games using Unity3D.
• Various computational optimization techniques are used to increase rendering FPS.
• I also had countless hours of experience with Autodesk 3ds Max (very similar to Autodesk Maya) prior to this, which jump started any kind of 3D projects I've had ever since.

Skills used: 3D modeling, Autodesk, Autodesk Maya