Hey, I'm Ashna 👋
I'm a software developer and data scientist based in Kansas City, Missouri. I specialize in building (and occasionally designing) websites, applications, and everything in between. Fun fact: This website was built with Next.js and hosted on Vercel!
About Me
Software Engineer and Data Science Specialist (M.S. Computer Science) specializing in applied machine learning, deep learning architectures, and high-reliability systems. Grounded in hands-on experience bridging real-time embedded software, predictive models, and cloud infrastructure, with a track record of converting complex data streams into high-impact, production-ready intelligence across healthcare, defense, and electric vehicle sectors.
Experience Highlights
- Driving Predictive Accuracy & AI Personalization: Delivered 80–90% diagnostic accuracy by pairing custom XGBoost and deep learning models with local LLMs for automated health recommendations, earning 3rd place at UMKC Hack-A-Roo.
- Accelerating Deep Learning Pipelines: Built GPU-accelerated training pipelines for computer vision models, streamlining large-scale image dataset processing for defense applications.
- Deploying Secure, Production-Grade Architecture: Engineered automated CI/CD cloud pipelines and secure EHR workflows to ensure compliance and high availability across regulated environments.
- Maximizing Powertrain Efficiency & Cost Savings: Programmed and integrated software for an updated traction motor system in heavy-duty electric trucks, reducing operational costs and improving vehicle efficiency.
Skills
HTML
Expert
CSS
Intermediate
JavaScript
Expert
React
Expert
Node
Expert
Python
Expert
AWS
Beginner
TensorFlow
Intermediate
Docker
Beginner
Java
Intermediate
SQL
Intermediate
Microsoft Azure
Beginner
C++
Intermediate
Arduino
Beginner
ESP32
Beginner
Pandas
Intermediate
Seaborn
Intermediate
Microsoft PowerBI
Beginner
SpringBoot
Beginner
Maven
Beginner
PyTorch
Intermediate
Projects
Cryptography Web App
A secure patient data storage solution combining AES, ECC, and PQC. Built with Docker and PostgreSQL, this graduate group project featured containerized database and app servers. This application is deployed on AWS Elastic Beanstalk.
Nine Men's Morris Game
A Python-based Nine Men's Morris board game with a Tkinter GUI, developed as a graduate group project.
Deep Learning Experiments
Four experiments using trained and custom AI models (e.g., Segformer, VGG, ResNet). Optimized for Google Colab GPU.
Racket Parser
This project implements a lexer and parser for simple mathematical language. The lexer tokenizes the input source code, and the parser processes these tokens according to predefined grammar rules. The project is designed to handle a language with basic constructs such as variables, arithmetic operations, and simple statements (e.g., assignment, input/output).
Graduate Teaching Assistant Job Board
A web application prototype for a Graduate Teaching Assistant Application form for the University of Missouri-Kansas City. Built with a MERN (MongoDB, Express, ReactJS, NodeJS) stack during my bachelor's program capstone project. The backend server is hosted on OnRender and the frontend is hosted on Vercel.
Cloud Builder
This repository contains a suite of services designed to automate the deployment of web applications using a microservices architecture. Each component of the project plays a specific role, from managing front-end interactions to handling backend operations and reverse proxy configurations. Microservices are managed by AWS.
AI-Powered Diabetes Health Coach
A Streamlit web app using XGBoost and LSTM models to classify diabetes risk levels. This 2024 Master's Data Science Capstone project won third place in the UMKC Hack-a-Roo competition. This application is deployed on Streamlit Community Cloud.
Mining Worker Safety Helmet IoT System
an IoT-enabled wearable device designed to enhance safety in hazardous underground mining environments. By integrating environmental monitoring sensors, location tracking, and real-time communication, this innovative helmet provides miners and management with critical safety data and alerts to improve situational awareness and response times. Uses an ESP32, Arduino Uno, and ReactJS for the frontend dashboard application.
Energy Company Customer Billing & Management System
a Java-based desktop application built with Swing for the GUI and integrated with Microsoft SQL Server via JDBC. It supports secure user authentication, dynamic customer record management, and automated invoice generation. The system includes modular UI components, customizable customer field editing, and external library integration (e.g., jdatepicker). Designed with extensibility and maintainability in mind, it demonstrates proficiency in Java desktop development, relational database integration, and event-driven programming.
Netflix Dataset EDA
Conducted an exploratory data analysis (EDA) of Netflix’s Movies and TV Shows dataset to uncover trends in content growth, genres, ratings, and regional distribution. Cleaned and visualized the dataset using Python (Pandas, Matplotlib, Seaborn), producing actionable insights into Netflix’s content strategy.
Neural Networks Implementation
A comprehensive neural network implementation featuring both from-scratch components and practical heart disease prediction models.
Student CRUD API
This project demonstrates a comprehensive implementation of a CRUD (Create, Read, Update, Delete) API with robust error handling, data validation, and comprehensive testing. It serves as a showcase of modern Java development practices and enterprise-level application architecture.
Get In Touch
If you want us to work together, have any questions or want me to speak at your event, my inbox is always open. Whether I just want to say hi, I'll try my best to get back to you! Cheers!
Say Hello