• Hello, I’m Angelo Karam.
• I’m interested in embedded systems, computer vision, natural language processing, and much more!
• I’m currently working on my Computer Science degree, along with minors in Mathematics and Physics at Sacramento State University.
• Looking for internships to get real-world programming skills that are not taught in school.
• How to reach me: angelokaram@gmail.com

A past research project I have worked on was titled "Assessing the Accuracy of Binary Star Detection Methods using Numerical Simulations of Star-Forming Clouds". The project entailed a series of programs created using Python and Dr. Aaron T. Lee's simulation data of star-forming clouds. These programs enabled the user to observe 3D star-forming systems (natal microcosm) from a random viewing angle, creating a 2D projection of the 3D data. These programs included sophisticated mathematical functions that were hand written into the Python program. We then used the programs and observed/noted that:

• True binary star systems are determined based on physics — their mutual gravitational interactions keep them bound to one another. The simulation data provided me enough information to calculate this.
• Perceived binaries systems are determined by finding the “smallest” apparent separation between stars in my projections. Lastly, I asked: Are the perceived binaries (the stars labeled as binaries by my programs) the actual binaries (those flagged as true binaries from the simulation data)? This question was ultimately answered with some statistical analysis of the generated data compared with the input simulation data.

There was an emphasis on science communication for this project. This meant that any and all aspects of the project had to be easily understandable by a non-scientific audience. A demonstration of this clear communication can be found in a poster I created with Dr. Aaron Lee, my mentor for the internship.

• This poster can be found in this repository, with the title "Astrophysics Research Poster" (here).

**The code for this project is set to private due to copyrighted contents. A draft for publishing to a scientific journal has been produced. Publishing has not occurred.

I am working under the guidance of a mathematics PhD to apply mathematical concepts, such as the Hebbian Learning Rule, the sigmoid function, and Boltzmann Brains to a Python programming project. This project aims to use genomic data and the handmade RBM to predict disease susceptibility.

I am currently working with a team of Software Engineers to build our client a ML model and simple UI that will grade students' homework/tests using Computer Vision and NLP. This project will save the client many hours each week in manual grading, which will allow them to focus more on curriculum and lecture planning. This project is currently in progress, but we project that a minimum of 10 hours a week of the clients’ time will be saved as a result of our well-trained model.

This project involved using the Logisim software to create a fully-functional CPU with working RAM, accumulator, TTY, register files, ALU, etc. All modules and wiring were done by hand. The aim of the project was to grasp the fundamentals of computer hardware and architecture.

A secure React web application to our client (Vendia). The goal of this project was to build a proof of concept distributed information system with suitable user interfaces that could be used by the FDA, pharmaceutical companies, and participating health care providers to exchange study data as it is produced. This was built with the information system on top of the Vendia Share platform, which supports data exchange through a distributed ledger and provides the primitives to control data flow (e.g., redaction). Led and collaborated with a team of seven peers to create all deliverables and the web app.

A simple minesweeper-like game built on a custom game engine. The project aim was to become familiar with Lightweight Java Game Library (lwjgl), OpenGL, JS, unique shaders, textures, OOP principles (SOLIDS), IO interrupts. The game and engine successfully completed and can be viewed on the above GitHub link.

An application that allows users to scan barcodes at retail stores to show price comparisons at other retail stores. This simple tool saves users money and cuts the time wasted trying to look up products on other websites. Project is for educational purposes, not complete, and backlogged.