Linda Dominguez

Linda Dominguez

Building intelligent solutions at the intersection of AI and software engineering

Scroll down
Linda Dominguez

Linda Dominguez

AI Engineer @ Jefferies Group LLC

dlinda@mit.edu

My Journey

I'm a Wellesley College and MIT alum. I studied Computer Science, Machine Learning, Data Science and Psychology related topics.

Currently, I'm an AI Engineer at Jefferies Group LLC contracted through Widenet AI (an AI consulting company), where I'm designing and developing a document extraction system. I'm leveraging my skills in TypeScript, React, and AI to build robust frontend and backend components while optimizing system performance and automating document processing workflows. I am also focused on enhancing user experience through effective UI design.

I'm actively seeking AI engineering opportunities in the fintech sector, where I can apply my expertise in machine learning, data science, and software development to create innovative solutions for complex financial challenges.

GitHub Stats

Experience

AI Engineer

Feb 2025 - Present

Jefferies Group LLC (via Widenet AI)

  • Partnered with Jefferies' CTO and engineering team to design and develop a document extraction system, reducing manual processing time and improving data accuracy.
  • Conducted tests to evaluate optimal approach for data input, comparing OCR, direct file input to model, or both.
  • Built UI and backend components using TypeScript and React, delivering seamless UX.
  • Diagnosed system issues, optimized performance, and automated document processing & data validation.

Data Scientist

July 2024 - Dec 2024

Fusion Alpha AI

  • Developed and optimized LSTM models to enhance ticker(stock) buy/sell signals, improving accuracy by 5%.
  • Created and upkept data pipelines using AWS, MongoDB, and S3 to source, process, and store data.
  • Automated data collection from various APIs and performed feature extraction for thematic analysis, informing targeted outreach strategies and improving CRM automation.
  • Used MongoDB database for storing historical data and conducting analytics on large datasets.
  • Developed and improved React-based system to automate email management using Gmail APIs, boosting operational efficiency by 30%.

Machine Learning Engineer Intern

Sep 2023 - Dec 2023

Biointerphase

  • Built random forest & NLP to predict bat population decline (97% accuracy) using timestamp & geolocation data.

Fellowship

June 2023 - April 2024

Breakthrough Tech AI @ MIT

  • Received hands-on training in machine learning algorithms; earned ML Foundations certification.

Projects

UFC Fight Tracker

Prediction Bet – UFC Fight Tracker IN PROGRESS

Full-stack web app that predicts UFC fight outcomes with real-time betting odds and sentiment analysis from Twitter. Uses ML to flag mismatches between sentiment and market odds.

React FastAPI MongoDB Python VADER
Ensemble Training Framework

Low-cost Ensemble Training IN PROGRESS

Developed a low-cost ensemble framework for CIFAR-10, using pruning, BatchEnsemble, and distillation to reduce training time and memory usage while maintaining high accuracy.

PyTorch CIFAR-10 Ensemble Learning Model Pruning
Crypto Sentiment Analysis

Sentiment Analysis on Cryptocurrency

Developed LSTM & GRU models for Bitcoin price prediction using tweet sentiment & historical data. Used Alpaca API & NLTK for data extraction and sentiment analysis.

LSTM GRU NLP Alpaca API NLTK
NY Botanical Gardens Classification

NY Botanical Gardens Classification

Built a multicategorical ResNet101V2 model to categorize 120,000+ images belonging to 10 classes of plant & animal specimen, with 97% accuracy. Top 15 out of 75+ national teams.

ResNet101V2 TensorFlow Keras Numpy
HealthAssess: AI Health Tool

HealthAssess: AI Health Tool

Developed web-based AI health assessment tool using live heart rate data, integrating GPT-4 API for real-time analysis & prediction. Designed for MakeHarvard 2024 AI competition.

Flask GPT-4 API Healthcare Python
Bat Population Predictive Model

Bat Population Predictive Model

Found, standardized, and merged 3 bat population datasets. Built random forest & NLP to predict bat population decline (97% accuracy) using timestamp & geolocation data.

Random Forest NLP Tokenizer Geographical Data
Cryptocurrency Price Prediction

Cryptocurrency Price Prediction

Trained multiple classification models (SGD, XGBoost, Random Forest, SVM, KNN) to predict cryptocurrency types, achieving up to 99.51% accuracy.

Alpaca API SGD Random Forest XGBClassifier Scikit-learn
Cleaning Website Contract

Cleaning Website Contract

Hired to create a cleaning website for a cleaning company in Manhattan. Helped business grow and gain customers.

Bootstrap JavaScript HTML/CSS Web Development

Courses

Computer Science

  • CS111 Computer Programming & Problem Solving Wellesley
  • 6.100L Introduction to CS and Programming using Python MIT
  • CS230 Data Structures in Java Wellesley
  • 6.009 Fundamentals of Programming MIT

Web Development

  • 6.9620 Web Lab MIT
  • 6.S063 Design for the Web: Languages and User Interfaces MIT
  • CS220 Human Computer Interaction Wellesley

Artificial Intelligence & Machine Learning

  • 6.390 Introduction to Machine Learning MIT
  • CS232 Artificial Intelligence Wellesley
  • 6.S191 Introduction to Deep Learning MIT

Data Science

  • 6.S079 Software Systems for Data Science MIT

Skills

Programming Languages

Python TypeScript HTML/CSS SQL R

Machine Learning & AI

TensorFlow Keras LSTM GRU ResNet Random Forest NLP Computer Vision XGBoost SGD Artificial Intelligence

Data Science

Pandas NumPy Matplotlib Seaborn Data Pipeline Development Feature Engineering Time Series Analysis API Integration Scikit-learn Jupyter Notebook Excel

Web Development & Tools

React Vue.js Flask Bootstrap Tailwind CSS JSON RESTful APIs Figma Git MongoDB Linux Object Oriented Programming Test Driven Development Algorithms

Get in Touch

Email Directly