JPMorgan Chase is the largest bank in the United States, and processes commercial loans for office, mixed use, industrial, and retail properties in local markets across the country. Students on this team will implement predictive analytics to develop models that forecast important aspects of the business process within the commercial loan division, with a focus on utilizing large language models, and identifying extreme events within a very large and complex data environment. All student team members will become JPMC 2024 Summer Interns, based in Jersey City, NJ, across the river from Manhattan, NY.
Predictive analytics utilizes data analysis, machine learning, artificial intelligence, and statistical models to identify patterns that predict future behavior. Students will develop a tool, models, and analysis to support operational improvement within the Commercial Loan division. The team will leverage the internal JPMC AI platform to support their work. They will focus on a wide range of AI techniques to address relevant use cases and questions from the internal stakeholders. We expect that there will be a particular focus on Large Language models, and their ability to identify the most effective and least effective conditions within a highly complex and data dense environment.
Identifying patterns of extreme events within the business process (both positive and negative) is an important first step in identifying and implementing improvements. New AI tools hold the promise of accelerating business process improvement within the organization.
Minimum Viable Product Deliverable (Minimum level of success)
- Develop an understanding of the current business process, products/services offering, client interaction standards, and technical challenges. Interview important stakeholders. Document most relevant use cases
- Literature review of all relevant techniques (AI, Machine learning, Statistical Forecasting, etc.), similar applications, patents, etc. and review current best practices within the sponsor’s organization
- Create basic models (any method) to demonstrate student understanding of the business environment and the feasibility of implementing predictive analytics within particular business process environments
- Implement large language models to demonstrate the efficacy of the method for addressing at least one of the Stakeholder use cases
Expected Final Deliverable (Expected level of success)
- Refine the initial models and analyze their performance in the most relevant use cases
- Validate conclusions
- Provide a roadmap for recommended next steps, evaluating options against their likelihood of success, and the potential benefit that they may yield
Stretch Goal Opportunities: (High level of success)
- Demonstrate Additional Use Cases
- Initiate a live predictive model within the business decisions process of one of the Stakeholders
Advanced Data Science and Modeling Techniques
Specific Skills: Applied project experience with Large Language Models and other applied AI techniques OR advanced coursework
Likely Majors: DATA, STATS, MATH, CS
Specific Skills: General skills in Data Science, good software development skills, and a willingness to quickly develop new technical skills as required for the project
EECS 281(or equivalent) is required
Likely Majors: DATA, CS
Specific Skills: General Programming skills, good software engineering practice and design, and a willingness to quickly develop new technical skills as required for the project
EECS 281 (or equivalent) is required
Likely Majors: CS, DATA, BBA/CS
Additional Desired Skills/Knowledge/Experience
- Successful team-based project experience. Excellent interpersonal skills
- Project Management utilizing Agile/Scrum
- Experience in business process analysis
- Interest in and general knowledge of Commercial Banking
- Practical experience implementing predictive analytics in a complex data environment
- Ability and desire to independently learn new technology skills as necessary for the project
- Experience implementing large language models, neural networks and self-supervised / semi-supervised learning models
Applied AI/ML Lead, Wholesale Loan Technology
Santosh is based in Jersey City, New Jersey and has been with the firm for over 4.5 years working on multiple NLP related use cases including but not limited to Search, Information Retrieval, Document Question & Answering, and LLM’s. Mr. Chikoti has 15 years of overall experience in Data Science, Machine Learning, and NLP areas mostly working on Customer Analytics use cases especially marketing data science covering problem statements like Customer Churn Prediction, Experience Personalization, and Up-Sell/Cross-sell Strategies, etc. Mr. Chikoti is passionate about – “Building value-driven ML/NLP products/services for business problems using science, data and technology” and looks forward to helping the MDP Team do the same. Mr. Chikoti has also filed for 6 US patents. Outside of work, he lives in Monroe Township, New Jersey with his wife and their daughter.
Executive Director, Wholesale Loan Technology
Todd is based in Chicago, Illinois and has been with the firm for over 24 years, portraying many roles from application development to multi-year program management involving architecture required to support numerous lines of businesses and products. Mr. Ippen is presently the Head of Loan Servicing, Digital Loans, and Collateral Technology. In addition, Mr. Ippen was responsible for the technical build and roll-out of the firm’s strategic loan platform to international markets in Europe, Latin America, and Asia. Outside of his day-to-day activities, Mr. Ippen is the JP Morgan Captain for University of Michigan recruiting, and is a member of two Chicago Technology Leadership Work streams (Recruiting & Global Innovation). Mr. Ippen is also the Chicago Tech Center University Recruiting Lead for all College Entry Level Software Engineering Programs (Internships and Full Time). In his free time, Mr. Ippen enjoys spending time with his wife and two children, traveling, boating, skiing, and attending various sporting activities.
Dr. Narayanasamy is a Professor in the Electrical Engineering and Computer Science Department. His research interests are at the intersection of computer architecture, trusted hardware, and program analysis, and his current focus includes computing systems for genome sequencing and health analytics. He received his PhD from the University of California, San Diego, advised by Brad Calder. Satish has spent several brief stints at Intel and Microsoft Research and co-founded a precision health start-up, Sequal Inc.
Project Meetings: During the winter 2024 semester, the JPMC team will meet on North Campus on Fridays from 2:00 – 4:00 PM.
Work Location: This project will take place on campus, with remote access to JPMorgan systems during Winter 24/Fall 24. All project team members are required to participate in the JPMorgan summer 2024 internship program based in the Jersey City, office. Jersey City is located across the Hudson River from Manhattan, and is on the New York City subway/PATH system.
Course Substitutions: CE MDE, ChE Elective, CS Capstone/MDE, DS Capstone, EE MDE, CoE Honors, IOE Senior Design, SI Elective/Cognate
Citizenship Requirements: Students selected for this team must have the right work in the U.S.A indefinitely, without sponsorship. All student team members are required to participate in JPMC 2024 Summer Internship program.
IP/NDA: Students will sign an IP and NDA agreement that is unique to JPMC.
Summer Project Activities: Summer 2024 Internships (in Jersey City, NJ) are guaranteed and required for all students who match to this project team. Every student who accepts an offer to join this project team must also participate in the summer internship with JP Morgan. The approximate dates of the Internship are June 3 – August 9, 2024.
- All project students must participate in the JPMorgan summer internship program, and will be based in Jersey City, NJ
- All students must
- Have the right to work in the United States indefinitely without sponsorship.
- Have cumulative University of Michigan GPA > 3.0
- Pass a background and drug test
MDP project offer letters will include required dates, compensation, and other conditions of the internship. Students will have 10 days to accept or reject their combined MDP/JPMC offer.