BMEII AI/ML Symposium, 2025
Presented research on transformer-based sentiment analysis in motivational language at the Biomedical Engineering and Imaging Institute, NY Academy of Medicine.
Harvard Hackathon, Top 5, 2025
Built Verus Defense AI, a media authentication dashboard for detecting deepfakes and verifying sensitive imagery integrity.
NYU Courant Institute, 2025
B.A. in Data Science, minor in Chemistry
02 / IN PRACTICE
Experience
Machine Learning Engineer Research Intern
ICAHN SCHOOL OF MEDICINE AT MOUNT SINAI · SUMMER 2025 – PRESENT
I build NLP pipelines on clinical and psychiatric data to study how language reflects motivation, agency, and mental health. My current work extends a transformer-based sentiment analysis pipeline to examine how motivational agency interacts with treatment outcomes in a 7-Tesla fMRI biofeedback trial for major depressive disorder. I’m currently writing this up as a manuscript.
Data Analytics Research Assistant
SOUTH ASIAN HEALTH AND RESEARCH CLUB, NYU SCHOOL OF GLOBAL PUBLIC HEALTH · MAY 2023 – JUNE 2024
I joined a cross-disciplinary team studying the real-world impact of “Roti,” a social media chatbot delivering dietary health education to South Asian American young adults. I handled survey data analysis and conversation metadata processing, including chi-square analysis, contributing to work published in JMIR Formative Research in 2024.
Data Science Tutor
CENTER FOR DATA SCIENCE, NYU · 2025
I tutored undergraduate students in Principles of Data Science I and II, covering Python, statistics, and applied ML concepts.
03 / BUILT TO LEARN
Projects
Ideas explored through code, research, design, and a willingness to make the first version.
CatCare
A deployed nutrition recommendation system built end-to-end in PyTorch, FastAPI, and React. Combines neural collaborative filtering with AAFCO nutritional compliance to generate personalized food recommendations for cats.
PowerNap
A machine learning pipeline that forecasts hourly grid carbon intensity to help reduce the carbon footprint of everyday computing tasks.
Low-Dimensional Temporal Coding of MT Neurons
An analysis of how visual cortex neurons encode motion direction through temporal firing patterns, not just firing rate. Applied PCA with permutation-based validation to single-unit electrophysiology recordings.
ResuMatch
A Streamlit app that optimizes resumes for specific job postings using NLP-based matching and scoring. Built end-to-end as a group ML course project.
Verus Defense AI
Harvard Hackathon Top 5: a media authentication dashboard combining deepfake detection, metadata analysis, and cryptographic verification for defense teams assessing sensitive imagery. I led the UX/UI design.
Assessing Professor Effectiveness
A statistical deep-dive into what drives professor ratings and whether gender plays a role. Used OLS regression and logistic classification on large review datasets.
Microtrend Radar
A geospatial pipeline detecting neighborhood-level fashion microtrends from resale listing data, using PostGIS, windowed SQL, and trend velocity metrics to surface hyperlocal signals.
04 / SHARED OUT LOUD
Presentations & Publications
Talks, articles, and ideas I’ve shared with others along the way.
PUBLICATION
Ali SH, Rahman F, Kuwar A, Khanna T, Nayak A, Sharma P, Dasraj S, Auer S, Rouf R, Patel T, Dhar B. Rapid, Tailored Dietary and Health Education Through A Social Media Chatbot Microintervention: Development and Usability Study With Practical Recommendations. JMIR Formative Research. 2024;8:e52032.
PUBLICATION
Morris LS, Beltran J, Kvamme T, Butler G, Adams A, Patel T, et al. Targeting the dopaminergic midbrain with precision 7-Tesla biofeedback training in depression: A proof-of-principle randomized controlled trial. Under Review.
PRESENTATION
Patel T. Investigating the Role of Sentiment in Motivational Strategies: Transformer-Based Models. Poster presented at the BMEII AI/ML Symposium, Biomedical Engineering and Imaging Institute, NY Academy of Medicine; 2025.
View Poster
PRESENTATION
Patel T. Investigating the Role of Sentiment in Motivational Strategies: Transformer-Based Models. Poster presented at the NYU Undergraduate Research Conference; 2025.
05 / CARE IN MOTION
CatCare
A nutrition recommendation system built for foster cats that combines personalized machine learning with nutritional standards.
I built CatCare because I couldn’t find a reliable, science-backed way to figure out what my foster cats should actually be eating. Most recommendations online are vague, brand-sponsored, or don’t account for a cat’s individual needs. So I built something that does.
CatCare combines neural collaborative filtering with AAFCO nutritional compliance standards to generate personalized food recommendations for cats.
01 / PROFILE
Start with the cat.
Enter a cat’s age, weight, health considerations, and other essentials. CatCare turns that individual profile into a nutrition-aware starting point.
02 / RECOMMEND
Learn, then verify.
The model learns preference patterns while a compliance layer checks each recommendation against AAFCO thresholds — personalization with a nutritional floor.
03 / TOOLKIT
Built to travel.
PyTorch · FastAPI · React · MLflow · Railway
04 / PRODUCT VIEWS
05 / WHAT’S NEXT
CatCare is a starting point. The long-term vision is a tool shelters can use to manage nutrition across large populations of cats with varying needs, and that individual owners can trust when they need real answers, not guesswork.
06 / THE TOOLBOX
Skills & Tools
The technical tools, creative practices, and systems I reach for when it’s time to make something real.
Languages
Python, SQL, R, Java, JavaScript, HTML/CSS
ML & Data Science
PyTorch, scikit-learn, TensorFlow, XGBoost, LightGBM, CatBoost, Optuna, MLflow, SHAP, Fairlearn, conformal prediction
NLP & LLMs
HuggingFace Transformers, SiEBERT, TextBlob, TF-IDF, GPT-4o annotation pipelines, sentiment and agency classification
Geospatial & Earth Observation
Google Earth Engine, Sentinel-2, Sentinel-1 SAR, CHIRPS, PostGIS, NDVI/EVI/NDWI
Data & Engineering
FastAPI, Streamlit, Docker, Railway, PostgreSQL, pandas, NumPy, SciPy
Frontend & Design
React, Framer, Figma
Tools & Platforms
Git, GitHub, Google Drive, PyCharm, Jupyter, VS Code
06 / OUTSIDE THE BRIEF
More Interests
The curiosities, communities, and small obsessions that keep my work connected to the wider world.
Travel
some of my favorite places i’ve been to!
CURRENTLY LISTENING
my current favorite album.
The Cats
MY PERMANENT KITTIES
Lassi
Nova
FOSTER ALUMNI
What I’m Reading
If Cats Disappeared from the World
Genki Kawamura
A quiet meditation on what we lose when the things we love are gone.
All About Love
bell hooks
The most honest thing written about what love actually requires of us.
My Year of Rest and Relaxation
Ottessa Moshfegh
Uncomfortable, funny, and more relatable than I’d like to admit.