Building intelligent solutions that bridge data and human needs through analytics and artificial intelligence.
I'm a data analyst and AI enthusiast originally from Tajikistan — a small, mountainous country in Central Asia. Growing up in a place where access to advanced technology was limited only fueled my curiosity.
I've always been fascinated by how things work — from the cosmos to computers — and that early passion led me to pursue a future in data, artificial intelligence, and systems design.
Achieving my dream of studying in the United States, I earned my degree from the University of Nebraska at Kearney. During that time, I built real-world tools — including a predictive model for healthcare billing and an AI-powered chatbot that integrates document ingestion, OpenAI GPT, and vector search.
I currently serve as a Graduate Assistant for the Office of Graduate Studies and Academic Innovation at UNK. In this role, I support academic program development, analyze student success data, assist in marketing efforts, and help enhance institutional effectiveness through data-driven reporting.
Used extensively for data cleaning, automation, and developing AI-based tools like my chatbot and healthcare billing error predictor.
Applied supervised learning and NLP models using libraries like scikit-learn and OpenAI APIs to solve real-world problems.
Built interactive dashboards and clear visuals using Python (matplotlib/seaborn), Tableau, and Excel to aid decision-making.
Developed full-stack AI applications combining GPT, vector databases (like Pinecone), and real-time user interfaces.
Worked with MongoDB and PostgreSQL to design efficient, secure, and scalable data storage systems for both structured and unstructured data.
Intelligent Campus Assistant for Students & Staff
This full-stack AI-powered chatbot was built as a capstone project to modernize student support at the University of Nebraska at Kearney. Designed for both students and staff, the chatbot can answer academic, administrative, and campus-related queries in real time using an intelligent backend pipeline that mimics how a real university help desk operates — but faster, smarter, and 24/7.
Reduced support load, improved student access to resources, and can be deployed as a campus-wide plug-in.
Predictive analytics tool for healthcare billing accuracy
This predictive analytics tool was developed to estimate the likelihood of medical billing errors in breast cancer treatment workflows. The hypothesis: The more procedures a patient undergoes, the higher the chance of billing errors. Our model analyzes procedure counts by stage and predicts error probabilities using custom-constructed datasets and statistical methods.
Found a significant increase in billing error probability when procedure count surpassed critical levels.
Have a project in mind or want to discuss potential opportunities? Feel free to reach out!