Master’s in Data Science: Capstone Project

On this page: Overview • Fall 2025 Capstone Project List • Fall 2024 Capstone Project List • Best Fall 2024 & 2025 Capstone Posters

Overview

CDS master’s students have a unique opportunity to solve real-world problems through the capstone course in the final year of their program. The capstone course is designed to help students apply their knowledge to practical situations and develop critical skills such as problem-solving and collaboration.

Students are matched with research labs within the NYU community and industry partners to investigate pressing issues, applying data science to the following areas:

Probability and statistical analyses
Natural language processing
Big Data analysis and modeling
Machine learning and computational statistics
Coding and software engineering
Visualization modeling
Neural networks
Signal processing
High-dimensional statistics

Capstone projects provide students with the opportunity to work in their field of interest and gain exposure to applicable solutions. Project sponsors, NYU labs, and external partners benefit from having a fresh perspective applied to their projects.

“Capstone is a unique opportunity for students to solve real-world problems through projects carried out in collaboration with industry partners or research labs within the NYU community,” says capstone advisor and former CDS Research Fellow Anastasios Noulas. “It is a vital experience for students ahead of their graduation and prior to entering the job market, as it helps them improve their skills, especially in problem-solving contexts that are atypical compared to standard courses offered in the curriculum. Cooperation within teams is another crucial skill built through the Capstone experience as projects are typically run across groups of 2 to 4 people.”

The Capstone Project offers organizations the chance to propose a project that our graduate students will work on as part of their curriculum for one semester. Information on the course, along with a questionnaire to propose a project, can be found on the Capstone Fall 2025 Project Submission Form.

Companies, nonprofits, and research labs! Interested in submitting a Capstone project proposal for students to work on? Please reach out to ds-capstone@nyu.edu.

Fall 2025 Capstone Project List

A Framework for Generative Music Models and Creator Tools with Ethical Licensing and Attribution
Agentic AI for Automating People Discovery, Outreach, and Dashboard Generation at Scale
Agentic AI Framework for Commercial Use Cases with MCP Compliance
Agentic AI Operators for Freight Forwarder
Agentic Health AI
AI Agent for Exploratory Data Analysis and Anomaly Detection
AI Assistant for Clinical Trial Protocol Design
AI dectection tools for improving the quality of online research
AI-Driven enhancement of Breast MRI interpretation using large-scale imaging and clinical data
AI-Powered Pharmacovigilance System for Oncology Drug Safety Monitoring
Assessing LLM Output Stability in Mental Health Applications
AssetOpsBench: Benchmarking Autonomous Agents for Industrial Asset Operations
Attention-Based Transformers for Transparent Survival Analysis
Audio-Visual Speaker Diarization Leveraging Foundation Models
Automating LLM-driven Counterfactual Data Generation
Autonomous Agent for Real-Time Issuer related Event Detection and Impact Analysis
Bibliometrics and LLMs for Transplant Knowledge Distillation
Bridging DNA and Protein language models for multi-species regulatory sequence prediction
Build a better resume screener
Cognition-Aware Human Modeling for Semi-autonomous driving
Context2SQL
Delivering Elite Football (Soccer) Analytics
Detecting and classifying species biodiversity in the waters of Barbados and the Eastern Caribbean
Developing methods for large-scale data integration through semantic mappings
Developing Multi-Modal Deep Learning Models for MRI Analysis
Do LLMs produce gender bias in hiring?
Exploring Perceptual Loss in Deep Learning-based MRI Image Reconstruction
Generative Pre-Training (GPT) for Structured EHR data
Improving Institutional Bitcoin Accumulation Strategies
Improving Tokenization by optimizing BPE
Machine Learning for Download Quality Detection
Machine Learning for Entity Linking
Model and predict videoconference communication experience quality using multimodal data
Modeling Human Decision Conflict with the Substrate-Prism Neuron
Multi-Agent Document Intelligence System
Multimodal Foundation Models for Stroke Severity Prediction
Overview | ACM ICAIF 2024 FinRL Contest
Predicting Progression to Chronic Pancreatitis Using Time-to-Event AI on Longitudinal CT Data
RAG framework and Agentic Health
Recalled Experiences of Death (RED): A Natural Language Processing Approach to Determining its Themes and Distinguishing it from other Human Experiences.
Shrinking human genes with generative modeling
Specialized AI Platform for Clinic Decision-making Support in Cancer Therapeutics
Strategic Funding Analytics: Predictive Modeling for EdTech Investment Opportunities
Tiny Movements, Big Insights: Deep Learning for Infant Motion and Emotion Recognition from Video
Transformer-based Causal Inference in the Real World
Using AI to Create Synthetic Commercial Data Sets

Fall 2024 Capstone Project List

A machine learning framework for identifying interacting DNA elements from multiomics data
A Non-biased Resume Screener in the Public Interest
Advanced AI Techniques for Dynamic Portfolio Management and Financial Forecasting
AI Tools for Journalists
AI-Driven Reimbursement Code Discovery for HealthTech Start-Ups
AI-powered deconvolution of RNA-sequencing data data
American Colleges and Universities As Social Networks
Audio-Visual Speaker Diarization Leveraging Foundation Models
Calculating Weights for Causal Inference using Wasserstein GANs
Deep learning models for cryptocurrency price analysis
Designed Validation Sets
Detecting Cognitive Impairment in Patient-Provider Communication
Developing Multimodal Detection Systems for Narratives Promoting Misleading Information
Developing Real-Time World Insight Reports with Machine Learning and LLMs
Digitizing P&ID into Industrial Knowledge Graphs
Dual-RAG: Advanced Financial Video QA with Multimodal RAG system
Enhanced Automatic Annotation of Spoken Adult-Child Interactions
Enhancing Large Language Models for Psychiatric Assessment
Estimation of a Latent Space Model for Strategic Network Formation
Hepatocellular Carcinoma Recurrence Prediction Radiomics Model
Improving genetic prediction of disease risk
Improving MRI Image Reconstruction via Generative AI
Information-constrained emergent communication
Large Language Models Encode Search for Clinical Evidence in Cancer Therapeutics
Learning to detect breast cancer utilizing all imaging modalities
Life Time Value (LTV) Model
Long-form Egocentric Video Understanding
Machine Learning-Driven Performance Monitoring in NYU’s DWDM Optical Transport Network
Model and predict videoconference communication experience quality using multimodal data
Optimizing Generative AI Modeling Stack for Sophisticated Financial Question Answering
Optimizing Kidney Composite Allocation Score Using Simulation Optimization
Political Partisanship and the US Judicial System
Power Grids and Markets: Quantitative Trading and Analysis
Predicting Agency Project/Program Success
Real-Time Interpretation and Translation for Low-Resource Languages
Relational Deep Learning and Graph Neural Networks for Anti Money Laundering
Scaling Satellite Imagery of Urban Streets via Automated Classification and OpenStreetMap
Solar Foundation Models
Spatial Morbidity and Co-morbidity in the U.S.
Tofu’s Candidate Matching Model
Understanding Distribution Shifts in Time
Use AI and math tools to do real estate appraisals
User-centric AI-models for assisting the visually impaired
Utilize large language model to analyze radiology report
Video representation learning

Best Fall 2024 and 2025 Capstone Posters

2024: Optimizing Generative AI Modeling Stack for Sophisticated Financial Question Answering

Student Authors: Haoyan Yang, Yixuan Wang, Ruxi Zheng, Zheng Tong

Project Mentors: Urjit Patel, Jessie Yeh, Chinmay Gondhalekar

2025: Detecting and classifying species biodiversity in the waters of Barbados and the Eastern Caribbean

Student Authors: Amaan Mansuri, Vishwa Raval, Shravan Khunti

Project Mentor: United Nations Development Programme (UNDP)

2023 Posters

Best Fall 2023 Capstone Posters

Multimodal NLP for M&A Agreements

Student Authors: Harsh Asrani, Chaitali Joshi, Tayyibah Khanam, Ansh Riyal | Project Mentors: Vlad Kobzar, Kyunghyun Cho

Partisan Bias and the US Federal Court System

Student Authors: Annabelle Huether, Mary Nwangwu, Allison Redfern | Project Mentors: Aaron Kaufman, Jon Rogowski

Best Fall 2023 Student Voted Posters

User-Centric AI Models for Assisting the Blind

Student Authors: Gail Batutis, Aradhita Bhandari, Aryan Jain, Mallory Sico | Project Mentors: Giles Hamilton-Fletcher, Chen Feng, Kevin C. Chan

Multi-Modal Foundation Models for Medicine

Student Authors: Yunming Chen, Harry Huang, Jordan Tian, Ning Yang | Project Mentors: Narges Razavian

Best Fall 2023 Student Voted Runner-Up Posters

Representational Geometry of Learning Rules in Neural Networks

Student Authors: Ghana Bandi, Shiyu Ling, Shreemayi Sonti, Zoe Xiao | Project Mentors: SueYeon Chung, Chi-Ning Chou

Medical Data Leakage with Multi-site Collaborative Training

Student Authors: Christine Gao, Ciel Wang, Yuqi Zhang | Project Mentors: Qi Lei

Fall 2023 Capstone Project List

A machine learning model to predict future kidney function in patients undergoing treatment for kidney masses
Advanced Name Screening and Entity Linking Using large language models
Automated assessment of epilepsy subtypes using patient-generated language data
Bringing Structure to Emergent Taxonomies from Open-Ended CMS Tags
Build Models for Multilingual Medical Coding
Building an Interactive Browser for Epigenomic & Functional Maps from the Viewpoint of Disease Association
Causal GANs
Designing Principled Training Methods for Deep Neural Networks
Developing predictive shooting accuracy metric(s) for First-Person-Shooter esports
Discovering misinformation narratives from suspended tweets using embedding-based clustering algorithms
Does resolution matter for transfer learning with satellite imagery?
Egocentric video zero-shot object detection
Evaluating the Capability of Large Language Models to Measure Psychiatric Functioning
Explanatory Modeling for Website Traffic Movements
Extracting causal political narratives from text.
Fine-Tuning of MedSAM for the Automated Segmentation of Musculoskeletal MRI for Bone Topology Evaluation and Radiomic Analysis
Foundation Models for Brain Imaging
Housing Price Forecasting – Alternative Approaches
Identify & Summarize top key events for a given company from News Data using ML and NLP Models
Improving Out-of-Distribution Generalization in Neural Models for Astrophics and Cosmology?
Knowledge Extraction from Pathology Reports Using LLMs
Leverage OncoKB’s Curated Literature Database to Build an NLP Biomarker Identifier
Measuring Optimizer-Agnostic Hyperparameter Tuning Difficulty
Medical Data Leakage with Multi-site Collaborative Training
Metadata Extraction from Spoken Interactions Between Mothers and Young Children
Multi-Modal Foundation Models for Medicine
Multimodal NLP for M&A Agreements
Multimodal Question Answering
Network Intrusion Detection Systems using Machine Learning
Online News Content Neural Network Recommendation Engine
OptiComm: Maximizing Medical Communication Success with Advanced Analytics
Partisan Bias and the US Federal Court System
Predicting cancer drug response of patients from their alteration and clinical data
Predicting year-end success using deep neural network (DNN) architecture
Prediction of Acute Pancreatitis Severity Using CT Imaging and Deep Learning
Preparing a Flood Risk Index for the State of Assam, India
Representational geometry of learning rules in neural networks
Segmentation of Metastatic Brain Tumors Using Deep Learning
Social Network Analysis of Hospital Communication Networks
Supporting Student Success through Pipeline Curricular Analysis
Transformers for Electronic Health Records
Uncertainty Radius Selection in Distributionally Robust Portfolio Optimization
Unveiling Insights into Employee Benefit Plans and Insurance Dynamics
User-centric AI models for assisting the blind
Using Deep Learning to Solve Forward-Backward Stochastic Differential Equations
What Keeps the Public Safe While Avoiding Excessive Use of Incarceration? Supporting Data-Centered Decision-making in a DA’s Office