Mortality Prediction (In progress)
AWS cloud pipeline for patient mortality predictions using synthetic clinical data.
About me
I am a Cum Laude economist from Universidad de Los Andes with a passion for data science and the burgeoning potential of machine learning. Through international competitions, research efforts, and work projects, I have gained proficiency in tackling technical problems and communicating data-driven insights.
Recent Work
AI-Powered Text Analysis
Power BI dashboard exploring AI perceptions in NYT articles using GPT-3.5 for classification
ML Architectures
Pytorch implementation of CNN, fully-connected network, and other basic ML architectures.
Automated Budget Enforcement (In Progress)
AWS CloudFormation template to revoke user permissions and terminate resources when a budget threshold is met.
Interpretability Benchmarks
Reversed-engineered Transformer models as a benchmark for Interpretability methods.
Attention Is All You Need
Transformer implementation based solely on the 'Attention Is All You Need' paper.
Comparing Health Interventions
Estimation of the cost-effectiveness of health interventions in Colombia and Nigeria.
ML Interpretability Challenges
Challenges for AI reserchers focused on repairing model components, auditing behaviors, and identifying manipulated inputs.
Alejo-Plotly
Wrapper for Plotly Express to visualize high-dimensional arrays and tensors.