Portfolio
Computer Science & Statistics Student | Aspiring Quant Researcher
I am a senior at California State University, Northridge (B.S. in Computer Science and Statistics, expected May 2026) focused on quant research, data science, and data analytics. I enjoy building reproducible data pipelines, developing statistical and machine learning models, and turning large datasets into practical decisions.
Slide horizontally to view projects.
Analyzed 350K+ proprietary entries to improve training efficiency and benchmarking quality for JPL flight-related projects.
Engineered features by combining internal and external sources, built reproducible visual analysis in Power BI/Seaborn, and presented actionable recommendations to division leadership.
Learn moreBuilding a real-time computer vision system with Python, OpenCV, and PyTorch to classify ASL gestures and convert prediction translations from text to speech output.
Built a custom ASL Sign Language Dataset using OpenCV and MediaPipe, developed a real-time gesture pipeline, trained classifiers on labeled hand-sign data, and evaluated performance with iterative preprocessing and augmentation to improve robustness.
Led a 10-student team to build a mentorship platform with Flask, React, and SQL for personalized support in Intro to Mechanical Engineering.
Built a course-restricted assistant for ME101 using a RAG pipeline so students can ask ChatGPT-style questions answered only from approved CSUN course materials. Added a quiz-generation tab that creates multiple-choice practice questions by topic from the same uploaded materials, and developed a personalized analytics dashboard that tracks quiz performance metrics to help students identify weak areas.
Extended Dijkstra's algorithm with dual edge weights in Java/Python to evaluate trade-offs in multi-objective routing.
Compared runtime and quality trade-offs against standard shortest-path methods and documented how multi-objective constraints affect route selection and complexity.
Developing computer vision models in Python and PyTorch to detect early-stage cocoa seed infection with EfficientNet architectures.
Applied image preprocessing and augmentation workflows to improve model performance and evaluated architecture-level trade-offs for early detection accuracy.
Implemented statistical computing methods in R for large-scale multiple testing and high-dimensional inference.
Built bootstrap resampling workflows from scratch to support Higher Criticism research and contributed to computationally efficient methods for modern statistical inference.
If you are interested in collaborating, feel free to reach out.