Working Paper

A comparative study challenging static lexicon-based models (VADER) in high-volatility markets. The proposed architecture utilizes Large Language Models (LLMs) to extract context-aware sentiment, coupled with a Multilayer Perceptron (MLP) to filter stochastic noise. The hypothesis focuses on isolating high-fidelity alpha signals that traditional linear regression models miss.

Originally designed as a predictive ML model for drought patterns in CDMX using hydrological data. Pivoted to a crowdsourced reporting system to bridge data gaps. Designed the data ingestion pipeline to validate user reports against historical meteorological norms.

Engineered a LightGBM-based predictive model leveraging multi-spectral satellite imagery (GEE) and NASA POWER meteorological data. Conducted rigorous dimensionality reduction across 44 distinct ecological indicators to classify global algal bloom patterns with high precision (AUC/F1 validation).

Simulated autonomous agent behavior using Python (Mesa) and Unity to analyze strategic decision-making dynamics. Designed reward-based optimization functions within constrained state spaces, applying Monte Carlo sampling to identify emergent Nash equilibrium patterns.