Coursework & Projects

Intermediate and advanced topics in unconstrained optimization. This course inspired my notes on the mathematical foundations of convex optimization.

A mathematically rigorous study of machine learning, including Mercer’s theorem and reproducing kernel Hilbert spaces for understanding support vector machines. My course project was later published at MLSP 2023.

A deeper understanding of linear algebra that inspired me to create a dedicated blog for my lecture notes.

An introduction to Bayesian statistics that helped me expand my reference notes with advanced probability distributions.

An introduction to R programming alongside statistical machine learning topics.

My final project explored “Explainability of Machine Learning Models using Co-operative Game Theory.”

My final project examined “Generative AI: Ethical Challenges and Prevention Strategies with Watermarking and Traceability.”

I built Tableau dashboards visualizing crime data and Airbnb listings in Chicago. For my final project, I developed an R Shiny app that helps Airbnb guests explore nearby crime rates, with filters for time ranges and crime categories.

Decision trees, Naive Bayes, KNN, regression, K-means clustering, DBSCAN, hierarchical clustering, support vector machines, and association rules.

Relational data stores, codebooks, schema diagrams, NoSQL graph, key-value, document, and wide-column stores, and MapReduce. My final project proposed a scalable national cancer-genomic database integrating these concepts.

Foundations of business and data intelligence, including decision support systems; descriptive, predictive, and prescriptive analytics; data types; and measurement scales.

Advanced distributed and cloud computing concepts: system architectures, process management, communication mechanisms, fault tolerance, security, and consistency models.

Fundamentals of large language models, including transformer architectures, tokenization, pre-training, fine-tuning, and prompting. Retrieval-augmented generation (RAG) connects models with external knowledge through document retrieval and context construction. Agentic workflows combine reasoning, planning, and tool use to carry out multi-step tasks.