Projects

1. Utilized divide & conquer and dynamic programming strategies to create an algorithm to figure out DNA sequence alignment. (Course: Analysis of Algorithms).
2. Created path finding algorithms (BFS, DFS, A*) in a 3D maze, mini-Go game utilizing game playing strategies (minimax), and used a MLP to classify written numbers using the MNIST dataset. (Course: Foundations of Artificial Intelligence)
3. Medaled in a Machine Learning Kaggle competition to make fashion recommendations. (Course: Machine Learning)
4. Implemented computer vision and digital processing techniques like: edge detection, image warping, texture analysis, and object classification using CNNs. (Course: Digital Image Processing)
5. Design and implemented static websites, RESTful API webapps hosted on Google Cloud, and Android application development. (Course: Web Technology)

Note: Some of the projects shared above only describe the project and do not share code, per class policy.

Database Experimentation Assistant (DEA) is an experimentation solution for SQL Server upgrades. DEA can help you evaluate a targeted version of SQL Server for a specific workload. Customers upgrading from earlier versions of SQL Server (starting with 2005) to more recent versions of SQL Server can use the analysis metrics that the tool provides.

To learn more, check out the documentation here.

As part of Azure’s Open Source Database Services offerings, a customer has the opportunity to monitor and tune their relational database using intelligent performance offerings, including Query Performance Insights and Performance Recommendations.

To learn more, check out the documentation or watch the Microsoft Ignite session.

For a detailed description and live demo of this project, check out this page.

Screenshots of my final graphics project are below.

This course explored the areas of bioinformatics, computational genomics, and modeling/simulating aspects of life. As part of my final project, I had to use an existing model on NetLogo and modify it for my own use. I was interested explaining how the immune system reacts to the presence of a pathogen (ie. learned responses vs initial exposures to pathogens) and decided to show a model of how the immune system fights pathogens. For an added level of complexity, I added in the factor of evolving pathogens and shifted my model to also deal with the evolving pathogens.

A screenshot of the final model can be seen below.

As part of my model, I included different types of pathogens (ie. Influenza A pathogens) with varying strength and mortality rates. I added in a variable to account for a pathogen that attacks the immune system and another variable to account for if a vaccine is available for the vaccine. These presets were automatically set based on which pathogen was selected. Another dynamic variable was the count of pathogens present in your system before simulating the model. A graph at the bottom would display the counts of white blood cells vs pathogen cells.

Through the course of a summer at the Scripps Translational Science Intitute, I was able to work on a some personal and team projects. The wesbite I launched that summer details the work I did and can be found here.

One of my large scale projects for that summer was to create a chi-squared test in Python to apply to GWAS (or Genome Wide Association Study). A GWAS is useful genomically because it tests a ‘control’ genome (for us, HG19) with a ‘variable’ genome to determine where the differences, or single-nucleotide polymorphisms (SNPs), occured. Since a GWAS determines if the allelic frequency between genomes is different, I added code to determine if this difference in allelic frequency between genomes is significant. If signifcant, then the SNP locations in the ‘variable’ genome could tell us something about diseases and illnesses, like cancer (the primary research area of our lab).