Skip to main content
Large Language Models

The Capstone Showcase represents the pinnacle of the student experience in the University of Chicago’s MS in Applied Data Science program. This event merges academic knowledge with practical, real-world challenges across various key research areas. This year, 37 capstone teams presented their projects, leading all phases of problem-solving with the support of faculty, staff, and industry experts. While projects spanned numerous industries, data types, and methodologies, all teams exhibited professionalism, pride, and commitment to excellence. The judges observed that the Showcase featured a diverse array of research-focused, industry-generated, and independent student projects spanning healthcare, marketing, supply chain, financial services, and more.

Paula Payton is the Head of the Center for Machine Intelligence and Data Science (MInDS) at Deloitte Research & Insights. She also happened to be a distinguished industry judge for the Autumn 2024 Capstone Showcase, her third year running. Impressed with the caliber of the graduate student presentations, Payton said, “Talk about pushing back the frontiers in applied AI…. talented students presented unbelievably original solutions to classic industry problems. That was exciting for me to see as a professional in the field and also someone who cares very deeply about data science and AI education.”

One of the standout projects was “Once Upon,” a storytelling app that uses AI to craft personalized stories for children based on their imagination and preferences. The team, consisting of Hima Cheruvu, Nia Gangar, Rishika Ravindran, and Christian Piantanida, shared their inspiration: “Our inspiration came from a moment of pure frustration and imagination. A few years ago, one of our team members’ then 4-year-old neighbor tried to get Amazon Alexa to tell her a story about her favorite princess going to swimming class. Alexa couldn’t quite deliver, and her neighbor was so frustrated that she started yelling at Alexa. That experience stuck with her—it showed how much children want to hear stories that are unique and personally meaningful to them, yet existing technology struggled to meet that need.” The student team wanted to create a tool that inspires creativity while empowering kids to be part of the storytelling process.

Learn more about the projects that won Best in Show below.

Inference Analytics | Multilingual Language Processing – Patient Facing Gen AI Information Retrieval Responder

Presenters: Yexuan Sun, Yahui Zhou, Maya Mireles Rios, Yun Yang

Faculty Advisor: Utku Pamuksuz

Robotics | SKYSEARCH: Generalized Object Detection and Search

Presenters: Zachary Farahany, Duncan Calvert, Joon Park, Mohammad Ayan Raheel

Faculty Advisor: Steve Barry

Chicago White Sox | Deriving Swing Value & Context from Biomechanics Data

Presenters: Hank Snowdon , Anthony Buti, Alexander Yeager, Devan Joshi

Faculty Advisor: Jeanette Shutay

Conversational AI | Once Upon

Presenters: Hima Cheruvu, Nia Gangar, Rishika Ravindran, Christian Piantanida

Faculty Advisor: Nick Kadochnikov

Museum of Illusions | Real Time & Predictive Analysis Process for an Immersive Entertainment Business

Presenters: Qinglin Chui, Nina Zhang, Bianca Gunawan, Syed Muhammad Hasan Zaidi

Faculty Advisor: Don Patchell

GPT for Healthcare | MediNotes: SOAP Note Generation through Ambient Listening, LLM Fine-tuning & RAG

Presenters: Yuki Leong, Melody Gao, Christina Song, Johnny Ji

Faculty Advisor: Utku Pamuksuz

GPT for Healthcare | R2Gen2: Finetuning Multimodal AI for Radiology

Presenters: Aayush Verma, Ashmita Mukherjee, Rishab Mohan, Leo Li

Faculty Advisor: Utku Pamuksuz

Honorable Mentions

GPT for Healthcare | Radipedia: Customized, Transparent, and Multimodal Assistant for Radiology 

Presenters: Lea Wu, Linda Ji, Vinezha Panca, Wyatt Meng

Faculty Advisor: Utku Pamuksuz

GPT for Healthcare | RadiGraph: Extracting Clinical Entities and Relationships from Radiology Reports into a Knowledge Graph Using LLMs

Presenters: Jiayi Deng, Ningxin Kong, Kaiheng Yao, Hongxuan Zhao

Faculty Advisor: Utku Pamuksuz

Conversational AI | InternationAlly by PropertyPilot

Presenters: Kshitiz Sahay, William DeForest, Yijing Sun, Daichi Ishikawa

Faculty Advisor: Nick Kadochnikov

Robotics | Innovative LEGO Design: Transforming Drone-captured 3D Model of Objects into LEGO Design Plan

Presenters: Phil Cooper, Nancy Gu, Thomas Shi, Peter Ye

Faculty Advisor: Steve Barry

UChicago Medicine | Prostate MR Image Segmentation with U-Nets and Transfer Learning

Presenters: Bender Russo, Junseong Park, Shubhankar Kumar, Wade Hou

Faculty Advisor: Batu Gundogdu

Inference Analytics | Medical Language Processing – LLM-based Radiology Annotation System

Presenters: Yezi Liu, Liuyi Pan, Kyu Sung Cho, Runyu Shang

Faculty Advisor: Utku Pamuksuz

Conversational AI | Trading Hero

Presenters: Kelvin Hsueh, Peter Fu Chen, Yaoning Yu, Nathan Chen

Faculty Advisor: Nick Kadochnikov

arrow-left-smallarrow-right-large-greyarrow-right-large-yellowarrow-right-largearrow-right-long-yellowarrow-right-smallclosefacet-arrow-down-whitefacet-arrow-downCheckedCheckedlink-outmag-glass