Junior Year is Here
September 22, 2018
Hi, I'm back.
It's been some time since I last made a blog post, more than a year in fact. I had said then that I'd try and post weekly or at least monthly but life happens and sometimes you come across other things that take priority. But that doesn't mean I completely stopped working on my website as a whole. I've been making a lot of changes and adding a lot of new things that I hope you'll enjoy. I've added a few new projects since and been through an entire year of new and exciting classes at Northwestern since my last post that I'd love to share with you guys.
Sophomore Year Classes
Before I get started with my junior year which gets going in just a few days, I want to look back at all the things I've learned in the past year of being a full-time student studying computer science. One of the thing I'm most excited about is the field of machine learning and AI in general. I took two class this past year, one called, "Into to Artificial Intelligence" and the other just "Machine Learning". I had always had an interest in the field but until I took these classes, it always seemed like something I'd never be able to grasp until many years later and had experience in a professional environment. It helped that at the same time as taking those classes, I was also taking the essential mathematics needed to really understand what the algorithms did. For things like linear algebra which has a lot of differnt terms and ideas that aren't really found elsewhere, I found this site which really helped me refresh and solidify the concepts useful to machine learning and AI. The visuals are excellent and it's something I wish I had when I was taking the class last year.
I've come out with two major projects since my last post that I'd like to share. The first, and more complex of the two, is called Auth and it is a facial recognition user authentication system that allows users to login to a website using just their face. This project was part of a class requirement for the Machine Learning class I took last year. There are two parts of the backend. One, which handles rendering the pages the user sees is built using NodeJs and the other is a Python backend that utilized Tensorflow to handle facial recognition. The facial recognition system is built off of a revolutionary new technique from a small team of Google researchers called FaceNet. A full write-up from this project can be found here and you can test the demo here. There may be a chance that the facial recognition backend is offline because it uses a lot of resources on my minimal server but I plan to upgrade soon to one that would also have a dedicated graphics card.
The second project that I've been working on more recently is called Grid and it's a very simple personal photo sharing app heavly inspired by vsco.co. I use it to show off some of my favorite photos from a few different sources like Instagram and unsplash.com. As much as I enjoy using these other services to share my photos, there's something special about having a way to do it yourself where you have full control. At this point, I don't have any plans of adapting the app for others, but if there's enough interest, I'd definitely consider it.