Academic Success Classification

The goal of this competition was to predict the academic risk of students in higher education. My model was able to achieve 83% accuracy locally, and I scored within the top 20% of competitors. You can view my Kaggle notebook here.

Explore More

Legal Garner v1

Legal Garner LLM is a legal language model designed to help my client handle common FAQs about the process of acquiring an attorney in an efficient and automated way. Built on the robust Legal BERT architecture from Hugging Face, Legal Garner LLM is a Retrieval-Augmented Generation (RAG) model tailored specifically to the legal domain.

Explore More

Generative Pre-trained Transformer

This GPT model is an implementation of a transformer-based language model using PyTorch. It uses key algorithms and technologies including multi-head self-attention, positional encoding, and feedforward neural networks. The model architecture consists of an embedding layer, several transformer decoder layers, and a final linear layer. Each transformer decoder layer combines self-attention and feedforward layers, normalized with layer normalization and regularized with dropout. The model is trained using the Adam optimizer and cross-entropy loss on a dataset, enabling it to generate coherent text by predicting the next token in a sequence.

Explore More

Questified

Executive dysfunction is a behavorial symptom that affects a person's ability to plan, organize, and carry out tasks. It can make it difficult to start and complete tasks, even when the person knows what they need to do.

Questify is an upcoming mobile app designed to help people with executive dysfunction overcome their challenges. The app uses gamification techniques to promote dopamine release when accomplishing daily tasks, which can help to improve motivation and focus. Questify is still in testing, but it has already shown promise in helping people with executive dysfunction regain control over their lives.

Explore More

ISIC 2024 Skin Cancer Diagnostic Model

Description is pending, competition in final phases.

Explore More

Project NOVA

I am working on the portion of UTD's NOVA that processes our new 4D Radar technology. I am creating a occupancy grid based off of the provided velocities and point cloud. (More info pending, start date was September 4th.)

Explore More

Software Development Internship

At CompuGroup Medical US, I am currently updating and developing documentation in Swagger for the Aprima Integrations team.

Consulting

As a consultant, I spend time budgeting, researching and creating NLP or analytical machine learning models to automate processes for small businesses and local clients.

AI Training

This job is self employment through Data Annotation Tech and my responsibilities are prompt engineering and QA testing code, content and systems produced by NLP models, while also ensuring the code and content produced is ethical, safe and ready for end-users.

Testimonials

"Laura is one of the most promising and dedicated programmers I've ever met. The passion she brings to her work is unmatched by our peers in this industry. I can confidently say there have been very few others I've seen on the same level of dediczation and professionalism as Laura. She's helped me find direction on my personal projects and her future teams are lucky to have her!"

- Jordan Reid, UTD NSBE Member