I'm a graduate in Applied Mathematics (BSc + MSc), with my Master's degree from CIMAT (Centro de Investigación en Matemáticas). I have solid training and hands-on experience in Natural Language Processing, Machine Learning, and Data Science. My background blends mathematical rigor with practical modeling skills, and I’ve worked on projects ranging from deep learning for text classification to scientific computing and Bayesian inference.
I enjoy tackling real-world problems using deep learning, probabilistic modeling, and modern NLP techniques. I'm also exploring applications of Large Language Models (LLMs), multimodal learning, and computer vision.
Always open to collaborations in interpretable AI, text modeling, or scientific ML.
🔍 Focused on: Natural Language Processing · Large Language Models · Deep Learning · Artificial Intelligence · Computer Vision · Scientific Machine Learning · Mathematical Modeling · Multitask Learning · Probabilistic Inference
🧠 Core ML & NLP
🔬 Scientific & Mathematical Computing
🔁 Bayesian Inference & Probabilistic Programming
📊 Programming, Scripting & Analysis
⚙️ Developer Tools & Workflow
🔬 PINNs Framework
Modular architecture for solving PDEs using Physics-Informed Neural Networks in several domains, including inverse problems and Bayesian inference via t-walk MCMC.
🧠 Multitask Tweet Classification with RoBERTuito and TF-IDF
Implements a multitask pipeline to predict gender and nationality from Spanish-language tweets using RoBERTuito and TF-IDF features. Trained with joint loss and evaluated using joint accuracy and F1. Based on the PAN Author Profiling dataset (CLEF 2017).
💳 Fraud Detection Model API
End-to-end solution for insurance fraud detection using WOE transformation and logistic regression scorecards. Includes preprocessing, model training, and a FastAPI service for real-time scoring with interpretable results.
📍 Rest-Mex 2025: Sentiment Analysis for Mexican Tourist Texts
Competition project for sentiment and theme classification in Mexican tourist reviews. Combines deep learning and classical NLP techniques for real-world destination profiling. Paper in preparation.
🧳 Tourist Opinion Mining and Text Analytics
Applies text preprocessing, exploratory analysis, and feature selection to thousands of tourist reviews from 10 landmarks. Uses TF-IDF + Chi² to identify discriminative terms across destinations and predict rating-based sentiment.
🤖 Advanced Deep Learning – CIMAT
Explorations in transformers, LoRA, diffusion models, and RNNs for time series forecasting and physical systems.
- ⚙️ Modular PINN framework for PDE solving in scientific applications (repo)
- 📧 Email: [email protected]
- 💼 LinkedIn: linkedin.com/in/ezautorres