My Story
My journey into technology began through FBLA & DECA, where I discovered my competitive spirit and passion. As a freshman, I achieved the ultimate goal of becoming a National Champion, followed by a 4th place finish at nationals during my sophomore year.
During my sophomore year, my life took a profound turn when I learned that my mom had been diagnosed with breast cancer. This personal challenge sparked something deeper within me—a drive. This led me to get back to the FBLA Nationals stage placing 4th at the end of my sophomore year Throughout the summer after my 10th grade I made an attempt to understand how technology could be used to make a real difference in healthcare and people's lives.
That pivotal moment led me to dive into research and discover the incredible world of machine learning and artificial intelligence. I've since built two CNNs (Convolutional Neural Networks), including a breast cancer prediction model that holds special meaning for me, and I'm currently working on another exciting project that pushes the boundaries of what I've learned.
Projects
Click on project boxes to take you to the GitHub Repository!
CIFAR-10 CNN
This was a basic Convolutional Neural Network, the first I made! This took the data set of CIFAR-10, which is a dataset containing a variety of categories, this was trained using a basic keras Sequential model with pooling, convolutional 2D layers, and dense layers, although the accuracy was extremely low I am still hoping to experiment more with the project and get that accuracy up!
CIFAR-100 CNN
This project uses TensorFlow and Keras to implement a Convolutional Neural Network (CNN) for image classification from the CIFAR-100 dataset. The model analyzes 32x32 color photos from 100 different categories, such as cars, animals, home goods, and landscapes. The CNN architecture consists of dense layers with batch normalization and dropout for regularization after progressive convolutional layers (16→32→64→128 filters). Classification on categories ranging from aquatic mammals to vehicles is achieved through training with 50,000 images using the Adam optimizer over 50 epochs. The implementation is perfect for multi-class image recognition tasks since it includes thorough evaluation with confusion matrix visualization to examine model performance across all 100 classes.
Breast Cancer Prediction Model
I engineered a comprehensive breast cancer prediction system using Python and scikit-learn that achieves nearly 90% accuracy in distinguishing between malignant and benign tumors, demonstrating the powerful intersection of healthcare and artificial intelligence in supporting critical medical decisions. Built predictive models using logistic regression on the Wisconsin Breast Cancer dataset (569 patient samples), analyzing 30+ tumor cell characteristics including radius, texture, perimeter, area, smoothness, compactness, and concavity measurements, while implementing comprehensive data preprocessing pipelines, comparative analysis frameworks testing multiple machine learning algorithms including decision trees and ensemble methods, and robust model evaluation using confusion matrices, precision, recall, and ROC analysis. Created interpretable decision boundaries and visualization tools for medical professionals, established feature importance rankings to identify most critical tumor characteristics, and showcased how machine learning can augment medical diagnosis to potentially reduce diagnostic errors and support healthcare professionals in making more informed treatment decisions. I do plan to use this for further researches and study the cancer cells for future projects where I can implement this network.
ML Cancer Detection App from Blood Samples
Analyzing factors affecting cancer diagnosis through blood sample data. Research-focused project exploring biomarkers and their correlation with cancer detection.
NLP + Finance: Algorithmic Trading
LSTMs, and BERT to analyze the sentiment of financial tweets. The ultimate goal is to predict long-term stock market trends, which can help us manage our wealth and anticipate potential booms and busts.
My Skills
My skills lie in computer science including machine learning with Python, web development with HTML, CSS and JavaScript, additionally my coursework has taught me other languages such as Java or SQL from my FBLA experiences
Programming Languages
Machine Learning & AI
Systems & Networking
Tools & Frameworks
Achievements
Recognition and awards from various competitions and conferences in technology and business
Network Design
National FBLA 2024 Conference
Jul 2024
Management Information Systems
FBLA Regionals
Feb 2025
Network Design
FBLA Regionals
Feb 2024
Network Design
State FBLA 2024 Conference
Apr 2024
Networking Infrastructures
FBLA Regionals
Feb 2025
Management Information Systems
National FBLA 2025 Conference
Jul 2025
Trebuchet
State TSA 2025 Conference
Apr 2025
Worked with a team to engineer a trebuchet that was able to throw golf-like balls at a distance of nearly 30+ ft effectively. I handled engineering design, testing, and maintenance.
Management Information Systems
State FBLA 2025 Conference
Apr 2025
Retail Merchandising Series
Area 2025 DECA Competition
2025
Connect With Me
Copyright © 2025 Viraat Nellutla. All Rights Reserved.