Hello, I'm

Alexis Harvey

PhD Student in Computer Science

Researching applied AI and LLM behavior at Louisiana State University, with a focus on how AI and extended reality can support medical training, decision-making, and accessibility.

View My Research Get In Touch
Alexis Harvey
Scroll
01

About Me

I'm a PhD student studying how Artificial Intelligence and Extended Reality can support decision-making, accessibility, and training in real-world settings.

I'm a PhD student in Computer Science at Louisiana State University, advised by Dr. Felipe Fronchetti in the RISE Lab and working in collaboration with LSU Health. My research interests include applied AI, machine learning, LLM behavior, computational social science, and civic technology. My current work looks at AI and extended reality in medical procedural training.

I earned my B.S. in Computer Science from LSU's Roger Hadfield Ogden Honors College with a concentration in Data Science and Analytics and a minor in Mathematics. As an undergraduate, I worked as a Student Software Developer at LSU's Center for Analytics and Research in Transportation Safety (CARTS), building full-stack applications over large relational datasets.

Beyond the lab, I'm interested in what happens when Artificial Intelligence meets the systems people actually use. That includes Artificial Intelligence for civic engagement, like making legislation, elections, and news bias easier for the public to understand; machine learning in sports, from valuing collegiate athletes to modeling performance; and intelligent devices for healthcare, such as prosthetics that monitor patient vitals. Across all of it, my goal is research that helps people make better decisions and reaches beyond the paper.

PhD CS Student, LSU
3 Active Research Projects
B.S. CS, Honors College
02

Research

Systematic Mapping Study

AI and XR in Medical Procedural Training

A PRISMA-aligned systematic mapping study on the joint use of AI and extended reality technologies in medical procedural training, co-authored with Dr. Felipe Fronchetti. Covers six databases (IEEE Xplore, ACM DL, ScienceDirect, SpringerNature, PubMed, Embase) with two-reviewer screening supplemented by a local LLM consistency pass, and maps the literature by clinical domain, AI and XR technology, evaluation type, and level of human intervention.

Systematic Review LLM-Assisted Screening LaTeX / ACM Python
Independent Research

Intelligent Prosthetics for Remote Patient Monitoring

A two-part research paper: a PRISMA-aligned systematic mapping study of AI in prosthetics, followed by the gap-driven design of a prosthesis that extracts vital signs (heart rate, respiratory rate, SpO2, cuffless blood pressure) from arterial pulse signals using a 1D CNN and sequence model with IMU-conditioned motion artifact handling. Built a parametric 3D prosthetic model in Blender via Python to document sensor placement at the brachial pulse site.

Deep Learning Signal Processing Blender / bpy Python LaTeX
Undergraduate Honors Thesis

The Agora: Civic Intelligence Platform

A full-stack civic intelligence platform combining live political data with LLM-driven analysis. Features LegisLens (RAG-based legal research over ChromaDB with a custom bias lexicon), a Media Bias Analyzer, an interactive D3 news map, an electoral data dashboard, and a Congress.gov bills scraper feeding legislative analytics.

React TypeScript FastAPI ChromaDB PostgreSQL Google Cloud Run
Watch Demo →
03

Experience

Aug 2026 – Present

PhD Student, Computer Science

Louisiana State University – RISE Lab

Conducting research on applied AI and LLMs under Dr. Felipe Fronchetti, in collaboration with LSU Health. Leading a systematic mapping study on AI and XR in medical procedural training and designing a VR user study.

Sep 2025 – Jan 2026

Student Software Developer

Louisiana State University – CARTS

Developed and tested full-stack applications using C#, JavaScript, SQL, and front-end frameworks. Managed large relational datasets and contributed to software support and enhancement for transportation safety research.

Nov 2024 – Dec 2025

Chapter President

Delta Zeta Sigma Chapter, Baton Rouge, LA

Led 400+ members, ran Executive Council meetings, planned events, managed budgets and contracts, and ensured chapter alignment with national standards.

Jul 2024 – Sep 2024

Legal Clerk Extern

19th Judicial District Court, Baton Rouge, LA

Collaborated with Judge Kelly Balfour and staff on legal proceedings, document preparation, and case research.

Aug 2022 – May 2024

Men's Basketball Manager

LSU Men's Basketball

Managed team statistics, assisted coaches with recruitment and footage gathering, maintained equipment, and supported player conditioning.

04

Technical Skills

Programming Languages

  • Python
  • TypeScript / JavaScript
  • C / C++
  • C#
  • Java
  • SQL
  • HTML / CSS

AI & Machine Learning

  • PyTorch
  • XGBoost
  • RAG Pipelines
  • LLM Integration & Evaluation
  • ChromaDB / pgvector

Frameworks & Platforms

  • React / Vite
  • FastAPI / Django
  • .NET
  • PostgreSQL / SQLAlchemy
  • Apache Spark / HDFS
  • Firebase / Google Cloud
  • Git & GitHub

Research Methods

  • Systematic Mapping Studies
  • User Study Design
  • Usability & Accessibility Evaluation
  • Data Analytics
  • LaTeX / Overleaf

Languages

English (Fluent) Spanish (Proficient) German (Beginner)
05

Featured Projects

Independent Project

NowIL

A longitudinal NIL valuation platform for collegiate athletes, currently in active development as a real product. Tracks and dynamically projects estimated NIL values from live in-game performance using Playwright data collection, a PostgreSQL/SQLAlchemy backend, PyTorch LSTM and XGBoost models tracked with MLflow, and a FastAPI service layer.

PyTorch XGBoost FastAPI PostgreSQL MLflow Playwright
Deep Learning

What-If Sports

A multi-task deep learning NIL valuation simulator for Division I men's basketball and a counterfactual companion to NowIL. Pairs a BiLSTM with attention against a multi-task Transformer with a shared encoder and separate classification and regression heads. XGBoost won on raw accuracy, while the Transformer achieved superior calibration (ECE 0.029).

PyTorch Transformers XGBoost FastAPI Plotly
Industry-Academic Project

Contract Wizard

AI-powered legal document platform built with Performance Contractors, featuring semantic search and LLM-driven clause summarization over a legal document repository with PDF ingestion workflows. Featured in WBRZ's coverage of LSU's AI showcase.

React TypeScript FastAPI PostgreSQL
News Coverage →
06

Let's Connect

I'm always open to discussing research collaborations, applied AI projects, or ways to build technology with real-world impact.