Professional Experience

Pricing Optimization - MIT Collaboration

Led a national-scale pricing optimization project for furniture category in collaboration with MIT, achieving 5-20% profitability improvement through strategic pricing strategies.

Impact: Deployed web application for nationwide implementation with measurable profitability gains.

Technologies: NumPy, Pandas, Scikit-Learn, Random Forest, IBM Cloud Pack, IBM CPLEX, SQL, DBeaver

Advanced ML for Clothing Pricing

Developed cutting-edge models including Feedforward Neural Networks (FNN) and LSTM networks for clothing category pricing optimization, incorporating Time Series Clustering for demand analysis.

Technologies: Keras, TensorFlow, Azure Data Studio, SQL Server, Matplotlib, Seaborn

Fashion AI Innovation

Proposed innovative strategy integrating GANs, Computer Vision, and social media data analysis for fashion design and trend forecasting.

Technologies: PyTorch, OpenCV

Facial Expression Recognition (FER)

Developed FER model for in-store customer analysis using transfer learning with Xception and Inception architectures.

Technologies: TensorFlow, PyTorch, Google Colab

Computer Vision Automation

Led development of CV-driven automation system using MobileNet transfer learning, integrated into API for bot automation.

Methodology: SCRUM

Visual Search on GCP

Led team of 4 in creating visual search demo on Google Cloud Platform with object detection and semantic segmentation.

Technologies: GCP, Docker, Git, Linux, Flask, SCRUM

Statistical Methodology Enhancement

Enhanced statistical methodology for data analytics and supply chain, improving business trial measurement accuracy.

Methods: Hypothesis testing, regression analysis, Bayesian methods

Time Series Clustering for Judicial Statistics

Implemented innovative unsupervised clustering algorithm for strategic human resource allocation based on time-series patterns in judicial data.

Evolution: Initially developed in SPSS, later migrated to R for enhanced analysis capabilities.

Technologies: SPSS, R, Microsoft Office

Text Classification for Federal Justice Census

Led development of sophisticated NLP model for "National Census of Federal Justice Impartation" to pre-classify federal offenses using machine learning algorithms.

Impact: Achieved accurate pre-classification of diverse federal crimes, enhancing census data analysis process.

Technologies: NLP, Machine Learning, Microsoft SQL Server

Faculty of Science Teaching

Served as dedicated professor at UNAM's Faculty of Science, designing and conducting diverse mathematical and statistical courses.