Upi Shanker

🏀

About Me

I'm Upi Shanker, a Computer Science honors student at Florida State University with a passion for building scalable, user-centered software. I specialize in full-stack development, AI automation, and data analytics, and have applied these skills through internships and research at organizations like St. Jude Children's Research Hospital, Voxel Labs, and the DeVoe L. Moore Center. Whether I'm building interactive data visualizations, deploying secure AWS-hosted web apps, or training AI models for real-world document processing, I focus on writing clean, impactful code. I'm also deeply involved in the tech community as VP & Website Coordinator of the Computational Neuroscience Society, Webmaster for ACM at FSU, and Outreach Chair for AITP. My goal is to continue growing as a developer and contribute to meaningful, mission-driven projects.

Skills

Python JavaScript TypeScript Java C C++ HTML CSS React NextJS ThreeJS Figma Flask Spring AWS SQLite PostgreSQL Data Structures Object -Oriented Programming Rest API Pandas NumPy Seaborn Matplotlib Exploratory Data Analysis Natural Language Processing Large Language Models Prompt Engineering Data Labeling Git Github Postman Documentation Teamwork Communication Research Problem Solving

Education

Florida State University

Bachelor of Science in Computer Science, Minor in Mathematics

Activities and Societies: University Honors Program, University Research Opportunity Program, ACM at FSU (Webmaster), South Asian Student Association, CompNeuroSociety (Vice President & Website Coordinator), AITP (Outreach Chair), Codeeducation

Honors & Awards: Dean's List, Vires Scholarship

Experience

  • Designing data-driven reports by querying the Epic EHR system to improve decision-making
  • Training a custom AI model that transforms natural language prompts into visual dashboards
  • Performing exploratory data analysis, automating workflows using Python and SQL to enhance reporting accuracy
  • Designed and developed a responsive, user-friendly website using HTML, CSS, and JavaScript, ensuring seamless navigation across devices and optimizing components for maintainability and consistent branding.
  • Integrated interactive elements such as form validation and real-time updates using JavaScript, enhancing functionality and user experience.
  • Deployed and secured the website with AWS services, leveraging S3 for hosting, Route 53 for DNS management, CloudFront for CDN, AWS Certificate Manager for SSL encryption, and integrated API Gateway with Lambda for backend functionality, ensuring high availability, performance, and security.
  • Developed modular and accessible front-end components using TypeScript and Next.js, enhancing user workflows
  • Integrated RESTful API endpoints to fetch and update inventory data, implementing robust error handling
  • Optimized front-end performance, reducing initial page load time by 50% through dynamic imports, image optimization, and caching strategies
  • Developed and published 5+ interactive Tableau dashboards focused on public finance and economic data
  • Cleaned and structured datasets from 10+ state and local agencies, increasing data accuracy by 30%
  • Collaborated cross-functionally with research teams and senior leadership to align dashboards with policy goals
  • Developed a Python script to categorize duplicate titles in a SQLite database, reducing manual data entry by 75%
  • Created a SQLite database using Natural Language Processing (NLP) and TF-IDF vectorization to automatically analyze and score academic papers
  • Developed and deployed a Flask-based web application to queries the academic scores database
  • Building AI-driven document processing workflows using Python, NLP, and computer vision to extract and analyze unstructured data in finance and real estate.
  • Implementing and evaluating AI models (LlamaIndex, DocLLM) for document classification and data extraction to optimize automation efficiency.
  • Designing AI-powered automation pipelines, streamlining data workflows, and developing analytical insights to enhance business processes.
  • Improved AI performance and accuracy by designing effective prompts and refining responses generated by LLMs
  • Conducted iterative testing and fine-tuning of LLM outputs by evaluating relevance, factual accuracy, and tone
  • Collaborated with cross-functional teams to structure RAG pipelines to boost response precision by 29%
  • Annotated and labeled video datasets to train and enhance the performance of Artificial Intelligence models
  • Developed and optimized high-quality code to implement machine learning models, ensuring scalability
  • Refined prompt strategies and retrieval mechanisms to improve chatbot response accuracy by 37% in LLM + RAG pipelines
  • Developed and automated data pipelines for gathering, cleaning, and analyzing large consumer datasets using Python, improving data processing efficiency and accuracy
  • Conducted exploratory data analysis (EDA) and sentiment analysis on customer reviews, leveraging Python libraries like Pandas and NumPy to uncover actionable insights and inform business strategies
  • Created and delivered data-driven presentations and interactive dashboards, effectively communicating complex insights to stakeholders and supporting strategic decision-making

Projects

Gradehub

GitHub

  • Developing an application for grade management using PostgreSQL and Spring Boot (Java) for the backend
  • Designed and implemented RESTful APIs to support user authentication and course management
  • Currently building an interactive, responsive frontend with Next.js, enabling seamless user experiences
Java PostgreSQL REST APIs OOP

FSU Crime Map

Map

  • Engineered a data pipeline to extract, clean, and process crime data from FSU’s crime website using Python
  • Developed interactive map that visualizes campus crime incidents, integrating geocoding and mapping libraries
  • Deployed the interactive crime map as a static web application on GitHub Pages, ensuring easy public access
Python Pandas Web Scraping

CompNeuroSociety Website

Website

  • Built a responsive multi-page website for a neuroscience society using HTML, CSS, and JavaScript
  • Implemented custom dark mode toggle with persistent user preferences for improved accessibility
  • Developed dynamic event listings and interactive features with modular JavaScript, enhancing user engagement
HTML CSS JavaScript UI/UX Design

Contact

Feel free to reach out via any of the following platforms: