Data Scientist
ML & AI Engineer · Chicago, IL
M.S. in Data Science. Passionate about AI in healthcare and building ML solutions that actually matter. Python, TensorFlow, PyTorch, AWS, Azure, GCP.
I'm a Data Scientist and ML Engineer who genuinely loves digging into data and figuring out what it's trying to say. I hold an M.S. in Data Science and I'm at an early stage of my career where I'm excited to keep growing, take on new challenges, and work alongside people who care about doing good work.
My biggest passion right now is AI in healthcare. I truly believe machine learning has a huge role to play in early diagnosis, clinical decision support, and making healthcare more accessible to people who need it most. It's the kind of problem that keeps me up at night in the best way possible.
I've had the chance to work on real problems at places like the U.S. Treasury, Microsoft, and Vanguard, where I got hands-on with everything from building data pipelines to training classifiers and analyzing large datasets. I'm looking for a role where I can keep learning, contribute meaningfully, and work on things that actually make a difference.
Real projects — from healthcare AI to clinical decision tools. Each one started with a problem I wanted to actually solve.
Predicts your cardiovascular risk trajectory over the next 20 years. You upload an ECG scan, punch in your cholesterol numbers and lifestyle habits — CardioChrono's CNN reads the waveform, weighs it against your clinical profile, and generates a personalized PDF report with risk charts and preventive recommendations. It's the kind of tool that should be in every GP's office.
Type your symptoms the way you'd describe them to a friend — SmartSymptom figures out what you actually mean. It auto-corrects spelling, fuzzy-matches against a clinical symptom library, estimates how serious the situation is, and produces a health summary PDF you can hand to a doctor. Built on Random Forest classification with SpaCy NLP doing the heavy lifting on natural language input.
A multimodal neural network that combines DICOM mammogram images with patient risk factors — age, BMI, smoking history, family history — to classify tumors as benign or malignant. The model uses a CNN for image features and merges them with clinical inputs for a more complete picture. SHAP values explain which factors drove each prediction, and a desktop GUI lets clinicians upload scans and download a full PDF risk report per patient.
A rigorous ML study on the UCI Breast Cancer Wisconsin dataset. 569 patients, 30 cell nucleus measurements, one question: benign or malignant? The Random Forest model is tuned with GridSearchCV, cross-validated for consistency, and the most diagnostic features — like radius mean and perimeter — are surfaced with importance plots. Clean, documented, reproducible. Exactly what you'd want before trusting a model in a medical context.
Google Scholar
hover & click to reveal research data
Whether you're looking to collaborate on a research project, explore an idea that's been sitting in the back of your mind, talk about a role, or just connect, my inbox is always open. I genuinely enjoy conversations that go somewhere unexpected, so don't hesitate to reach out about anything.
Based in Chicago, IL. Excited about AI healthcare, ML, data analytics, and any team building things that genuinely matter. Open to remote, hybrid, or on-site.