Melad Mayaar

About Me

Hello there 👋

I am Ahmad Melad Mayaar, I go by "Melad," and I'm obsessed with optimizing life, whether it's through writing efficient code, hitting personal bests at the gym or scoring goals. I currently work full time as a Systems Engineer with a tech company in the city of Anaheim, CA. I graduated from Berkeley, with a B.A. in Data Science and minor in Computer Science in May of 2024. I spent my college days cleaning big datasets, testing software, building and deploying Machine Learning models and coding AI driven Pac-Man, but outside of that, you could find me weightlifting, hanging out with friends, or on the soccer field.

Career and research wise, I am particularly interested in:

- hardware–software co-design

- System AI

- Software Engineering with a great focus on Data Analytics

- Effective Deep learning HCI for models not limited to LLM

I am on the hunt of full time employment and research as well as exciting projects in the fields mentioned above, if you got any cool ideas, Let’s chat!

Contact: You can contact me via email at ahmadmeladmayaar@gmail.com or you can connect with me through LinkedIn (linked below). I'm looking forward to communicating with you!

Resume and Links

You can connect with me through some of the links listed below.

LinkedIn Profile

GitHub Profile

Resume

Skills

These are my top skills that I have gained through research, internships, and classes.

Web Development: HTML5, CSS, JavaScript, PUG, PostgreSQL

Data Science: Pandas, Web-Scraping, Data Gathering, Matplotlib, Seaborn, Plotly

Programming: Python, Java, C, C++, RISC-V, PostgreSQL, MongoDB

Developer Tools: Git, Visual Studio Code, PyCharm, IntelliJ, JupyterNotebook, Colab

Languages: Fluent in English, Farsi, Dari, & Pashto, Beginner in Arabic

Machine Learning: Feature Engineering, One-Hot Encoding, Sklearn, Regressions, Data Modeling, Hypothesis Testing,Clustering

Projects

The following are some projects that I have worked on through my internships and classes

Spam/Ham Email Classifier:

In this project, I was given a large real world dataset, and I was asked to classify the emails as spam or ham. I performed feature engineering on text data, utilized sklearn libraries for data processing, model fitting, and validation, and achieved 91% accuracy on the test set for the spam classifier. Additionally, I generated and analyzed precision-recall curves to assess the model's performance and make necessary improvements.

Exploring Cook County, Illinois Housing

In this project, I conducted an in-depth investigation of the Cook County property tax system, utilizing a data set of over 500,000 records, to analyze property valuation practices. Utilized Pandas, feature engineering, machine learning (one hot encoding) and Linear Algebra techniques to analyze data and identify disparities, highlighting reduced taxation for wealthier homeowners and increased taxation for low income households.

Designing a CPU

This project was the capstone project of CS61C where I used logisim-evolution to implement a 32-bit two-cycle processor based on RISC-V. During the CPU design process, I had several tasks to accomplish. Such as working on creating an Arithmetic Logic Unit (ALU) to handle calculations and logical operations. I also developed Register Files to store and manage data efficiently. The Data Path was another crucial component that I worked on, ensuring smooth data flow within the processor. In addition, I tackled the intricate task of decoding instructions, making sure the processor could understand and execute them properly. Finally I integrated Memory modules into the design, allowing the processor to interact with data stored in memory as well as the write back operations, ensuring that processed data was correctly stored.

N-Gram Viewer

Implemented a Java-based browser tool motivated by Google’s Ngram Viewer to analyze the historical usage of words in English texts throughout the history. Added cool data visualization feature in the browser tool, allowing users to interpret word usage trends through interactive charts and graphs.

Software Engineering Intern with Cerevox AI

During my internship as a software engineer at CereVox AI, I joined a small team of three developers to develop a top-notch, scalable custom parser. Our main goal was to build a parser capable of efficiently managing millions of customer requests. The parser's key task is to convert unstructured files into CereVox's specific format, making sure the data is consistent and standardized. We also made sure our parser could smoothly convert data from the CereVox format to any format a customer might need.

Avatar: The Way of Tiles

Lead group capstone project creating a video game engine that uses a randomized algorithm in Java to generate unique interactive 2D worlds in which the user can engage with the game using the GUI. Features random map generation, keyboard input, save and load structures, random encounters, various avatar skins, and unique seeds for corresponding unique worlds. Implemented a robust engine within the video game, enabling realistic object interactions, collision detection, and dynamic simulations for enhanced gameplay experiences.

Data Science Intern - UC Berkeley Data Discovery

As a Data Science intern, I worked alongside a team of six Data Scientists on developing a predictive model to predict lead water service lines across the nation. My role involved collecting data, scraping websites, building predictive models, and designing both the front end and back end of the User Interface. As the User Interface team leader, I employed the MapBox API to develop an informative and searchable parcel map. At the end of this endeavor, we were honored with the Data Science Insights Award from UC Berkeley, a recognition that really highlighted the importance of what we had accomplished.

Tailor-Friendly Software (Side Project)

I'm currently immersed in an exciting side project/startup, with a mission to create tailor-friendly software simplifying the management of customer clothing sizes and styles. This endeavor involves crafting a user-friendly interface that enables tailors to effortlessly manage customer data, including measurements and clothing types like Punjabi, Thobe, and more. To bring this vision to life, I'm delving into the world of PostgreSQL (psql) for data storage and exploring diverse software tools for both the back end and the user-friendly front end. My deep commitment to utilizing my Data Science and Software Engineering skills to serve my community and generate a positive impact drives this initiative.

Hello World! I'm Melad.

I follow excellence not success.