Miguel Diaz Email and Phone Number

Data Scientist | Neuroscientist | Machine Learning PractitionerI am a data scientist with a decade of research experience and a strong foundation in machine learning, statistical analysis, and predictive modeling. My work combines technical expertise in Python (Pandas, scikit-learn, PyTorch) with domain knowledge from neuroscience and data-driven problem-solving across diverse industries.Selected Projects and Achievements:Retail Lending Risk: Built and deployed a machine learning model to predict loan defaults, improving AUC by 28% through feature engineering and preprocessing of 300,000+ records. Model deployment utilized Google Cloud, FastAPI, and Docker.Stroke Risk Prediction: Optimized a novel classification model, increasing recall by 42% and AUC by 30%. Deployed the model in a containerized Flask application.Mental Health in Tech: Analyzed 230,000+ survey records using SQL and Python, uncovering key trends in mental health within the tech industry.Police Violence Analysis: Investigated fatal police shootings, creating interactive visualizations with Plotly and Matplotlib to reveal actionable insights.Mushroom Classification: Developed a neural network classifier in PyTorch using transfer learning with ResNet-18 for species identification via computer vision, achieving a performant model through advanced techniques.I specialize in feature engineering, model optimization, and data visualization, and I bring hands-on experience in deploying machine learning solutions to production using tools like Docker and cloud services. My analytical skills are complemented by strong storytelling through visualizations, enabling data-driven decisions.Beyond data science, my background in neuroscience adds depth to my understanding of complex systems, providing a unique perspective when tackling intricate datasets.My journey into machine learning began with a simple question: “What if I mastered the skills that once intimidated me?” As a neuroscientist, I leaned more toward biology, but I was fascinated by data and computation. Now, my work spans across domains—taking a problem-solving, domain-agnostic approach to data science. Whether in retail, healthcare, or beyond, I’m eager to tackle challenges that can be solved through predictive modeling and machine learning.When I’m not coding or exploring data, you can find me playing basketball, running, and enjoying the arts (fun fact: I started as an Acting major at NYU before shifting to Psychology and Neuroscience at UT Austin). Music-lover. I’m also an avid home chef.