Ethan Jordan

Backward Propagation Neural Network

I wanted to really understand how machine learning works, so I challenged myself to build a neural network from scratch using Python. This wasn’t just a class assignment—I spent a whole quarter digging into the math and code behind backpropagation, and by the end, I had a working model that could actually learn from data.

Along the way, I got hands-on with:

• Feedforward and Backpropagation: I wrote the code to make predictions and then update the network’s weights using calculus and gradients. Debugging this was tough, but it taught me a lot about how real AI works.
• Activation Functions: I experimented with ReLU, Sigmoid, and Tanh to see how they affect learning, and learned how important these choices are for real-world results.
• Preventing Overfitting: I tried out regularization and dropout, which helped my model generalize better—something I know is crucial in both research and industry.
• Matrix Math & Python Libraries: I got comfortable with NumPy and vectorized operations, which made my code much faster and more efficient.

This project gave me a real appreciation for the challenges (and excitement) of building AI systems. I learned to break down complex problems, test my code, and keep improving until it worked. I’m excited to bring this kind of curiosity and persistence to future research or software teams.

Mattercam Project

I love theme parks and data, so I built a dashboard that streams live video from Disneyland and shows up-to-the-minute ride wait times—all in a fullscreen, modern interface. This was my way of combining my interests in UI design, automation, and a little bit of fun.

🛠 What I Built & Learned:

• Custom Tkinter UI: I designed a clean, easy-to-read interface with color-coded ride times, so anyone (even my non-techy friends) could use it.
• Web Scraping: I used BeautifulSoup and Requests to pull live data from queue-times.com, which meant I had to figure out how to parse real-world, messy HTML.
• Live Video Streaming: I integrated a YouTube webcam feed using yt-dlp and VLC, which was a fun challenge in media handling and Python bindings.
• Real-Time Updates: The app refreshes data every minute and shows a timestamp, so you always know what’s happening in the park.

This project taught me how to bring together different technologies and think about the user experience, not just the code. I also learned how to troubleshoot when things break (which they did, a lot!). I’m proud of how it turned out, and I’d love to keep building tools that make data more accessible and engaging.

AVAC Tennis Class Automation

Working as a tennis scheduler at AVAC, I saw firsthand how much time was wasted on repetitive enrollment tasks. Instead of just complaining, I decided to do something about it: I built a Python automation tool that now saves our staff hours every month.

The project started simple—just scraping class rosters and enrolling students—but I kept adding features based on what my team needed. Now, it checks attendance, flags students for review, and even helps with tricky cases using CSV files and prompts.

🛠 How I Made It Work:

• Python + Selenium: Automated the browser to log in and handle the clunky Club Automation interface.
• Pandas & CSVs: Compared player lists and attendance logs to make sure no one slipped through the cracks.
• Real-Time Prompts: The tool asks for human input when something looks off, so we don’t make mistakes.
• Easy Setup: I wrote a batch script to set up the environment and keep credentials safe, so anyone on staff can use it.

What started as a personal time-saver is now a tool the whole department relies on. This project showed me how much impact software can have in the real world, and how important it is to listen to users and keep improving your work.

BinaryTree Class (Python Data Structures)

Original Author: An Lam
Edits and Implementation: Ethan Jordan
Date: 11/01/2023

I wanted to really understand how data structures work under the hood, so I took on the challenge of implementing a binary search tree in Python. I wrote the code, tested edge cases, and even added an interactive menu so users could explore different tree traversals.

🛠 What I Focused On:

• Core BST Operations: Insert, search, and delete—making sure everything worked smoothly, even for tricky cases like deleting the root.
• Tree Traversals: Built in-order, pre-order, and post-order methods, and learned how recursion and iteration can both be used.
• User Interaction: Added a simple text menu so anyone could try out the tree and see how the algorithms work in real time.

This project helped me build a strong foundation in algorithms and Python class design. It also taught me to think about how to make technical concepts accessible to others—something I hope to keep doing in research or teaching roles.

Python Computer Inventory Manager

Author: EJ
Date: 10/13/2024 (Updated 12/4/2024)

I built this command-line app to help keep track of computers for a small IT team. It started as a way to organize Linux and Windows machines, but quickly grew into a full inventory system with a database, input validation, and a user-friendly menu.

🛠 What I Learned & Built:

• Python + SQLite3: I set up a persistent database so nothing gets lost, even if the program closes.
• OOP Design: Used class inheritance to handle different types of computers, which made the code much cleaner and easier to expand.
• Input Validation: Made sure only valid IP addresses could be entered, which saved a lot of headaches later.
• Interactive Menu: Built a simple interface so anyone could add, update, or remove computers without needing to know Python.

This project was a great way to practice combining OOP, databases, and real-world problem solving. It also showed me how much users appreciate clear instructions and reliable tools—something I want to keep focusing on as I grow as a developer.

RTSP 4-Camera Viewer

Author: Ethan Jordan
Date: December 2024

I developed a professional multi-camera monitoring application that displays up to four RTSP camera feeds simultaneously in a clean, modern interface. This project combines my interest in video streaming technology with practical UI/UX design for security and surveillance applications.

🛠 Technical Implementation:

• VLC Integration: Used python-vlc bindings to handle RTSP streams with optimized caching and network settings for stable, low-latency video feeds.
• Custom Tkinter UI: Built a borderless window with a custom title bar, draggable interface, and responsive 2x2 grid layout that adapts to different screen sizes.
• Advanced Window Management: Implemented fullscreen toggle, window dragging, and keyboard shortcuts (F11 for fullscreen, Escape to exit) for professional user experience.
• Stream Optimization: Configured VLC with specialized parameters for live streaming including network caching, clock synchronization, and audio/video optimizations.

This application demonstrates my ability to work with multimedia libraries, create professional desktop applications, and solve real-world problems in video surveillance and monitoring. The clean interface and smooth performance make it suitable for security operations, event monitoring, or any scenario requiring multiple camera feeds.