Faculty Recommendations Now Open
Empower your students. Expand their opportunities.
What is the CRAFT Talent Network?
The CRAFT Talent Network connects high-potential students with full-time roles, exclusive internships, workshops, conferences, and networking events offered through CRAFT and its industry partners.
Industry partners include: Bank of America, BNY Mellon, Capgemini, Charles Schwab, Goldman Sachs, IBM, Park Ave Finance, Prudential, Vanguard, Wells Fargo and more.
Who Is Eligible?
Students must be recommended by a CRAFT-affiliated faculty member and meet one of the following criteria:
Currently or previously involved in a CRAFT-funded research project
Involved in summer or ongoing research projects that are fintech-related
How to Recommend a student
Email [email protected]
CC the student(s) you are recommending
State that you are recommending them for the CRAFT Talent Network
Once received, CRAFT will send the student(s) an enrollment link. Your name will be listed as the recommending faculty member.
Questions?
Contact us at [email protected]. Thank you for your continued support of CRAFT and your students.
Current and former students
The following students have worked alongside principal investigators on NSF-funded I/UCRC CRAFT Fintech Research Projects.
Hicham Abarkha
CRAFT Research Project: The Role of Wholesale CBDCs in Financial Stability: Models, Risks, and Global Case Studies
Educational Background
M.S. in Business Analytics at Rensselaer Polytechnic Institute, class of 2025
B.S. in Business Analytics dual with Economics at Rensselaer Polytechnic Institute (RPI), class of 2024
Bio
Hicham Abarkha is pursuing a Master’s in Business Analytics with a concentration in Quantitative Finance at Rensselaer Polytechnic Institute, graduating in May 2025. He previously earned a dual Bachelor’s degree in Business Analytics and Economics.
Hicham has developed strong expertise in data analysis, statistical modeling, and economic research. During his internship with the New York State Department of Health, he worked with Medicaid datasets, creating data visualizations and reports to support critical policy decisions. His research on wholesale Central Bank Digital Currencies (CBDCs) explored systemic risks, where he collaborated with professors and presented detailed analytical reports.
Beyond his academic achievements, Hicham gained hands-on experience as an Operations Intern at Amazon, where he led initiatives that resulted in significant cost savings and improved operational efficiency. Additionally, as an IgniteU Fellow, he worked with industry stakeholders to address challenges in disruptive technologies, further sharpening his problem-solving and consulting skills.
Hicham is passionate about using data-driven insights to solve complex problems in healthcare and finance, with a focus on creating innovative and impactful solutions.
Research Interests
Risk Analytics
Data-Driven Policy Making
Financial Innovation
Healthcare Analytics
Publications
The Role of Wholesale CBDCs in Financial Stability: Models, Risks, and Global Case Studies. (In Progress)
Contact
Yash Tanaji Bhosale
CRAFT Research Project: AI Compliance Officer
Educational Background
M.S. in Data Science, Stevens Institute of Technology
B.E in Information Technology with Major (Honors) in Data Science
Bio
Yash Tanaji Bhosale is an M.S. student at Stevens Institute of Technology specializing in large language models (LLMs), retrieval-augmented generation (RAG) and applied artificial intelligence systems. His work focuses on integrating embedding models, vector databases and advanced retrieval architectures to build scalable legal search, monitoring and reasoning systems across U.S. federal and state regulations.
Bhosale's research centers on information retrieval optimization, statistical validation of AI outputs, uncertainty quantification and real-time regulatory change detection.
He is particularly interested in building domain-specific LLM applications that combine structured data, regulatory text and intelligent retrieval pipelines to create reliable, explainable AI systems. His broader goal is to bridge advanced AI research with real-world infrastructure, developing solutions that are technically rigorous, commercially viable and impactful.
Research Interests
LLMs and RAG
Statistical Modeling
RegTech
Data Analysis
Publications
Ayush Jadhav, Prathamesh Govilkar, Yash Bhosale, Sahil Gamre, Deepali Kadam, “Empowering new Mothers: A Holistic platform for Women's Mental Health Enhancement Post - Childbirth”
Published - International Conference on Advances in Computing, Communication and Intelligence Systems (ICACCIS 2024)
Ayush Jadhav, Prathamesh Govilkar, Yash Bhosale, Sahil Gamre, Shubham Jadhav, Deepali Kadam “Unveiling the Past: A Holistic Approach to Rescuing Historical Texts through Advanced Image Analysis and AI” Published - ResearchGate March 2024
Contact
Andrew Carranti
CRAFT Research Project: Extending, Simulating and Scaling Decentralized Exchanges Made by Automated Market Makers
Educational Background
M.S. in Financial Engineering, Stevens Institute of Technology
B.S. in Quantitative Finance, Stevens Institute of Technology
Bio
Andrew Carranti is a M.S. in Financial Engineering student at Stevens Institute of Technology working under Professor Zachary Feinstein, Professor Ionut Florescu, and Professor Ivan Bakrac on the development of Automated Market Makers (AMMs) for traditional financial markets. His work for CRAFT has been primarily focused on AMM fee structure development and market simulation.in healthcare and finance, with a focus on creating innovative and impactful solutions.
Research Interests
Asset Price Prediction
Behavioral Game Theory
Quantum Finance Modeling
Publications
Tejas Appana, Andrew Carranti, Arjun Koshal, Adam Moszczynski and Matthew Thomas. “Gaussian Processes for Implied Volatility Estimation” (Accepted by NCUR 2024)
Contact
Zhi Chen
CRAFT Research Project: AI Compliance Officer
Educational Background
Ph.D. Candidate, Financial Engineering, Stevens Institute of Technology (August 2021 - Present)
M.S. in Financial Engineering, Stevens Institute of Technology (August 2019 - May 2021)
Bio
Zhi is currently focused on the intersections of fintech, large language models and ESG. His research aims to enhance financial decision-making and asset-pricing models. His notable work includes the creation of FinMem, a collaborative multi-agent system that uses synthesized large language models to optimize financial strategies from multiple sources. He is also exploring hierarchical algorithms to extract significant factors from ESG datasets.
Beyond research, he shares his expertise as a Python programming instructor, designing educational experiences that foster a deeper understanding of data analysis and modeling in finance.
Research Interests
Fintech
Large Language Model Agent System
Information Retrieval
ESG
Publications
Wang, Dan, Zhi Chen, Ionuţ Florescu, and Bingyang Wen. "A sparsity algorithm for finding optimal counterfactual explanations: Application to corporate credit rating." Research in International Business and Finance 64 (2023): 101869.
Cao, Yupeng, Zhi Chen, Qingyun Pei, Nathan Lee, K. P. Subbalakshmi, and Papa Momar Ndiaye. "ECC Analyzer: Extracting Trading Signal from Earnings Conference Calls using Large Language Model for Stock Volatility Prediction." In Proceedings of the 5th ACM International Conference on AI in Finance, pp. 257-265. 2024.
Cao, Yupeng, Zhiyuan Yao, Zhi Chen, and Zhiyang Deng. "CatMemo@ IJCAI 2024 FinLLM Challenge: Fine-Tuning Large Language Models using Data Fusion in Financial Applications." In Proceedings of the Eighth Financial Technology and Natural Language Processing and the 1st Agent AI for Scenario Planning, pp. 174-178. 2024.
Cao, Yupeng, Zhi Chen, Qingyun Pei, Fabrizio Dimino, Lorenzo Ausiello, Prashant Kumar, K. P. Subbalakshmi, and Papa Momar Ndiaye. "RiskLabs: Predicting Financial Risk Using Large Language Model Based on Multi-Sources Data." arXiv preprint arXiv:2404.07452 (2024).
Yu, Yangyang, Haohang Li, Zhi Chen, Yuechen Jiang, Yang Li, Denghui Zhang, Rong Liu, Jordan W. Suchow, and Khaldoun Khashanah. "FinMe: A Performance-Enhanced Large Language Model Trading Agent with Layered Memory and Character Design." arXiv preprint arXiv:2311.13743 (2023).
Yu, Yangyang, Zhiyuan Yao, Haohang Li, Zhiyang Deng, Yupeng Cao, Zhi Chen, Jordan W. Suchow et al. "Fincon: A synthesized llm multi-agent system with conceptual verbal reinforcement for enhanced financial decision making." arXiv preprint arXiv:2407.06567 (2024).
Li, Yang, Yangyang Yu, Haohang Li, Zhi Chen, and Khaldoun Khashanah. "TradingGPT: Multi-agent system with layered memory and distinct characters for enhanced financial trading performance." arXiv preprint arXiv:2309.03736 (2023).
Contact
Nan Cui
CRAFT Research Project: Federated Learning for Fairness-aware and Privacy-Preserving Financial Risk Assessment
Educational Background
Ph.D. in Computer Science, Stevens Institute of Technology (SIT), Hoboken, NJ (08/2020 - Present)
M.S. in Computer Science, Stevens Institute of Technology (SIT), Hoboken, NJ (01/2019 - 05/2020)
M.S. in Electrical Engineering, University of Massachusetts Lowell (UMass Lowell), Lowell, MA (08/2015 - 12/2017)
B.E. in Electrical Engineering and Automation, Northeast Electric Power University (NEEPU), Jilin, China (08/2011 - 07/2015)
Bio
Nan Cui is a Ph.D. student in Computer Science at Stevens Institute of Technology, working under Professor Yue Ning. She has a strong background in Computer Science and Electrical Engineering. She has worked on several projects, including the development of a federated learning framework tailored to support fairness-aware Graph Convolutional Networks (GCNs) and a fair and efficient active learning framework for designing label-efficient algorithms.
Research Interests
Graph Deep Learning
Federated Learning
Fairness in Machine Learning
Publications
Metric-Fair Active Learning. Jie Shen, Nan Cui, and Jing Wang. In Proceedings of the 39th International Conference on Machine Learning, 2022.
Applying Gradient Descent in Convolutional Neural Networks. Nan Cui. In Proceedings of the 2nd International Conference on Machine Vision and Information Technology, 2018.
History of Research Projects
Fair Graph Neural Networks in Federated Learning
Individual Fairness in Active Learning
Contact
Fabrizio Dimino
CRAFT Research Project: AI Compliance Officer
Educational Background
M.S. in Financial Technology and Analytics, Stevens Institute of Technology (August 2023 - May 2024)
Bio
Fabrizio Dimino is a passionate researcher at the intersection of Artificial Intelligence and Finance, focusing on the transformative potential of Multi-AI Agent Systems in Fintech. He is particularly interested in automating financial workflows, enhancing decision-making, and designing robust evaluation methodologies for AI-driven systems.
Research Interests
Multi AI Agent Systems
Fintech
Publications
Cao, Yupeng, Zhi Chen, Qingyun Pei, Fabrizio Dimino, Lorenzo Ausiello, Prashant Kumar, K. P. Subbalakshmi, and Papa Momar Ndiaye. "RiskLabs: Predicting Financial Risk Using Large Language Model Based on Multi-Sources Data." arXiv preprint arXiv:2404.07452 (2024).
Contact
Conor Flynn
CRAFT Research Project: DeFi Data Engine (May 2022 - May 2023)
Educational Background
Incoming Ph.D. Student in Computer Systems Engineering at Rensselaer Polytechnic Institute
Bachelor of Computer Science with a minor in Math at Rensselaer Polytechnic Institute, Class of 2023
Bio
Conor Flynn is an incoming Ph.D. student at Rensselaer Polytechnic Institute studying computer systems engineering. Having studied programming since middle school, Conor has since participated in various clubs, coding competitions, and leadership positions revolving around his love for computer science.
Throughout his professional journey, Conor has since worked as an intern as well as an entrepreneur. In 2019, he worked his first professional position as an Information Security Intern at PDX Inc., helping to secure their networking systems. In 2022, Conor was a Computer Science Intern at Convergint Technologies LLC, working on computer vision modeling for their camera systems to help improve their existing infrastructure. His entrepreneurial ventures started in 2020 with the founding of his first company Quantify Enterprises LLC. Throughout his duration working as its sole proprietor, Conor was contracted by various financial professionals and institutions to help improve their existing quantitative trading systems and strategies. Following his success at Quantify, he partook in the founding of ScaleTrade LLC with two colleagues in 2022, aiming to help traders improve their returns through the generation of unique and insightful technical data.
Conor also has done undergraduate research under the supervision of Professors Kristin Bennett and Professor Oshani Seneviratne starting in May of 2022 through May of 2023. During this time, he worked on a programming language and data agnostic, low latency data engine referred to as the DeFi Data Engine. This engine aims to improve the data collection and distribution for the lab for future research.
Research Interests
Low Latency Programming
Computationally Agnostic System Designs
Quantitative Finance
Deep Learning
Applied Mathematics
Publications
Adams, Kacy, Fernando Spadea, Conor Flynn, and Oshani Seneviratne. "Assessing Scientific Contributions in Data Sharing Spaces." arXiv preprint arXiv:2303.10476
(2023).
Contact
Michael Giannattasio
CRAFT Research Project: Risky Business? Deep Dives into DeFi
Educational Background
Rising Senior at Rensselaer Polytechnic Institute with a major in Mathematics
Bio
My name is Michael Giannattasio and I am a rising senior at Rensselaer Polytechnic Institute. I have a major in Mathematics specializing in Operations Research with a minor in Economics of Quantitative Modeling. I have been working on the CRAFT project Risky Business? Deep Dives into DeFi from the Summer of 2022 to the present. I have been focused on user-level representation of transactional data and discovering novel user-clustering methods, specifically across multiple datasets, to apply to multiple DeFi ecosystems' users at once. In order to display and facilitate the visualizations of these user clusters along with survival analysis results, I have been developing the DeFi Survival Analysis Toolkit with authored documentation located here.
Research Interests
Novel Clustering Methods
Risk Analysis
Data Science
Machine Learning
Artificial Intelligence
Publications
DeFi Survival Analysis: Insights Into the Emerging Decentralized Financial Ecosystem (Accepted by ACM-DLT); Characterizing Common Quarterly Behaviors in DeFi (Accepted by MARBLE 2023)
Contact
Satvik Gurjar
CRAFT Research Projects: AI Compliance Officer & Assessing the Risks of Alternative Data
Educational Background
M.S. in Financial Technology and Analytics | Stevens Institute of Technology
BTech. in Computer Engineering | Vishwakarma University ( India)
Bio
I am currently pursuing my Master’s in Financial Technology and Analytics at Stevens Institute of Technology. My academic journey has covered a wide spectrum of disciplines, including Risk Management, Statistical Learning, and Pricing and Hedging, providing me with a rigorous mathematical foundation for modern finance.
I am particularly drawn to Risk Management because it represents the ultimate challenge of quantifying uncertainty. To me, it is the bridge between theoretical models and market reality; I find it fascinating to build frameworks that safeguard capital by identifying hidden correlations and managing the volatility inherent in global markets.
During my studies, I have been deeply involved in two key research projects under CRAFT that define my approach to the industry. The first, AI Compliance Officer, is a compliance-focused initiative where we leverage Generative AI and Natural Language Processing to automate and simplify complex regulatory hurdles for SMBs. By building intelligent systems to parse through dense compliance documentation, we aim to make high-level legal adherence accessible to smaller firms. My second project involved a deep dive into Alternative Data, where I analyzed the risks and alpha-generating potential of non-traditional datasets. I focused on the complexities of data cleaning, signal decay, and the ethical implications of integrating unstructured data into institutional financial models.
My background in Computer Engineering provides me with the technical infrastructure to execute these complex financial strategies. Rather than just understanding the theory, I have the coding and analytical skills to build the systems that drive them.
Research Interests
Model Risk & Agentic AI in Financial Systems
Market Microstructure & Algorithmic Stability
FinTech
Publications
Satvik Gurjar, Chetna Patil, Ritesh Suryawanshi, Madhura Adadande, Ashwin Khore, Noshir Tarapore, “Mental Health Prediction Using Machine Learning”, IRJET, (2022), pp: 1-5.
“Assessing the Value and Risks of Alternative Data in Cross-Sectional Equity Return Prediction” (In Progress)
Contact
Inwon Kang
CRAFT Research Project: Risk Mitigation in Cross-Platform Decentralized Finance
Educational Background
Ph.D. Candidate, Computer Science, Rensselaer Polytechnic Institute (Current)
M.S., Computer Science, Rensselaer Polytechnic Institute (May 2022)
B.S., Computer Science, Rensselaer Polytechnic Institute (May 2020)
Bio
I am a current Ph.D. student at Rensselaer Polytechnic Institute, working under Professor Oshani Seneviratne. I work mostly with Python for data/analysis oriented projects. I am also familiar with javascript-based frameworks. During my undergraduate and master's program, I focused on explainable machine learning using shallow models. My master's project was on collecting survey data from participants and building explainable ML models to try to explain the results. During this time, I also dabbled in NLP and built an enhanced pipeline of existing state-of-art model for Preference Ellicitation task using GNNs and coreference parsers. My Ph.D. focus is on blockchain systems and integration of machine learning with blockchains.⠀⠀⠀
Research Interests
Blockchain Interoperability
System Design
Deep Learning
Natural Language Processing
Publications
Dependency and Coreference-boosted Multi-Sentence Preference Model. Farhad Mohsin, Inwon Kang, Yuxuan Chen, Jingbo Shang and Lirong Xia. (DLG-AAAI’23)
Learning to explain voting rules. Inwon Kang, Qishen Han, Lirong Xia (Extended Abstract, AAMAS 2023)
Analyzing and Predicting Success in Music. Inwon Kang, Michael Manduluk, Boleslaw Szymanski. (Scientific Reports)
Blockchain Interoperability Landscape. Inwon Kang, Aparna Gupta, Oshani Seneviratne. (IEEE BigData Distributed Storage Workshop 2022)
Landslide Likelihood Prediction using Machine Learning Algorithms. Vasundhara Acharya, Anindita, Inwon Kang, Thilanka Munasinghe, Binita KC (IEEE BigData 2022)
Crowdsourcing Perceptions of Gerrymandering. Benjamin Kelly, Inwon Kang, Lirong Xia. (HCOMP 2022)
Learning Individual and Collective Priorities over Moral Dilemmas with the Life Jacket Dataset. Farhad Mohsin, Inwon Kang, Pin-Yu Chen, Francesca Rossi and Lirong Xia. (MPREF-22)
Making group decisions from natural language-based preferences. Farhad Mohsin, Lei Luo, Wufei Ma, Inwon Kang, Zhibing Zhao, Ao Liu, Rohit Vaish and Lirong Xia. (COMSOC-21)
Research projects
Contact
Md. Saikat Islam Khan
CRAFT Research Project: Efficient, Private, and Explainable Federated Learning for Financial Crime Detection
Educational Background
Ph.D. Candidate, Computer Science, Rensselaer Polytechnic Institute (August 2023 – May 2028)
B.Sc. in Computer Science and Engineering (February 2015 – January 2020)
Bio
I am currently a Ph.D. student at Rensselaer Polytechnic Institute, working under Professor Oshani Seneviratne. My research activities span a broad spectrum of applications in machine learning, computer vision, and data science. During my undergraduate studies, I developed various classification models to classify medical images. My Ph.D. focus is on federated large language models and explainable federated learning. Through my involvement with CRAFT, I introduced Federated Learning for Relational Data (Fed-RD), a novel privacy-preserving federated learning framework specifically developed for financial transaction datasets that are partitioned both vertically and horizontally across parties. Fed-RD strategically employs differential privacy and secure multiparty computation to ensure the privacy of training data. I now aim to improve this project by enhancing the explainability of the black-box model.⠀⠀⠀⠀⠀⠀⠀⠀⠀
Research Interests
Distributed Systems
Federated Large Language Model
Explainable Federated Learning
Health Informatics
Publications
Fed-RD: Privacy-Preserving Federated Learning for Financial Crime Detection MSI Khan, A Gupta, O Seneviratne, S Patterson - arXiv preprint arXiv:2408.01609, 2024
A differentially private blockchain-based approach for vertical federated learning L Tran, S Chari, MSI Khan, A Zachariah, S Patterson… - arXiv preprint arXiv:2407.07054, 2024
Accurate brain tumor detection using deep convolutional neural network MSI Khan, A Rahman, T Debnath, MR Karim, MK Nasir… - Computational and Structural Biotechnology Journal, 2022
Water quality prediction and classification based on principal component regression and gradient boosting classifier approach MSI Khan, N Islam, J Uddin, S Islam, MK Nasir - Journal of King Saud University-Computer and …, 2022
MultiNet: A deep neural network approach for detecting breast cancer through multi-scale feature fusion SI Khan, A Shahrior, R Karim, M Hasan, A Rahman - Journal of King Saud University-Computer and …, 2022
Using fused Contourlet transform and neural features to spot COVID19 infections in CT scan images A Nur, MSI Khan, MK Nasir - Intelligent Systems with Applications, 2023
Proposing a hybrid technique of feature fusion and convolutional neural network for melanoma skin cancer detection MM Rahman, MK Nasir, M Nur-A-Alam, MSI Khan - Journal of Pathology Informatics, 2023
BlockSD‐5GNet: Enhancing security of 5G network through blockchain‐SDN with ML‐based bandwidth prediction A Rahman, MSI Khan, A Montieri, MJ Islam, MR Karim… - Transactions on Emerging Telecommunications …, 2024
Machine learning and deep learning-based approach in smart healthcare: Recent advances, applications, challenges and opportunities A Rahman, T Debnath, D Kundu, MSI Khan, AA Aishi… - AIMS Public Health, 2024
Federated learning-based AI approaches in smart healthcare: concepts, taxonomies, challenges and open issues A Rahman, MS Hossain, G Muhammad, D Kundu… - Cluster computing, 2023
Biorthogonal wavelet based entropy feature extraction for identification of maize leaf diseases B Mazumder, MSI Khan, KMM Uddin - Journal of Agriculture and Food Research, 2023
Contact
Maruf Ahmed Mridul
CRAFT Research Project: Smart Encoding and Automation of Over-The-Counter Derivatives Contracts
Educational Background
Ph.D. Student, Computer Science, Rensselaer Polytechnic Institute (August 2022 - Present)
M.S., Computer Science, Rensselaer Polytechnic Institute (August 2022 - December 2024)
B.Sc. In Computer Science and Engineering, Shahjalal University of Science and Technology (April 2014 - November 2018)
Bio
Maruf Ahmed Mridul is a third-year Ph.D. student in Computer Science at Rensselaer Polytechnic Institute (RPI). His current research focuses on automating the process of converting financial contracts into machine-readable formats using Large Language Models (LLMs) and Retrieval Augmented Generation (RAG).
Before joining RPI, Maruf was an Assistant Professor of Computer Science at Shahjalal University of Science and Technology in Bangladesh. He also worked as a software engineer at Samsung R&D, contributing to the development of features for wearable devices and robotics applications. His work reflects a blend of academic inquiry and practical experience in addressing complex computational challenges.
Research Interests
Large Language Models
Data Mining
AI
ML
Blockchain
Decentralized Systems
Publications
Mridul, M. A., Chang, K., Gupta, A., & Seneviratne, O. (2024, October). Smart Contracts, Smarter Payments:
Innovating Cross Border Payments and Reporting Transactions. In 2024 IEEE Symposium on Computational
Intelligence for Financial Engineering and Economics (CIFEr) (pp. 1-8). IEEE.
Kang, I., Mridul, M. A., Sanders, A., Ma, Y., Munasinghe, T., Gupta, A., & Seneviratne, O. (2024, May). Deciphering
Crypto Twitter. In Proceedings of the 16th ACM Web Science Conference (pp. 331-342).
Sharma, A. S., Mridul, M. A., & Islam, M. S. (2019, September). Automatic Detection of Satire in Bangla
Documents: A CNN Approach Based on Hybrid Feature Extraction Model. In 2019 International Conference on
Bangla Speech and Language Processing (ICBSLP) (pp. 1-5). IEEE. [Best Paper Award]
Sharma, A. S., Roy, T., Rifat, S. A., & Mridul, M. A. (2021). Presenting a Larger Up-to-date Movie Dataset and
Investigating the Effects of Pre-released Attributes on Gross Revenue. arXiv preprint arXiv:2110.07039.
Sharma, A. S., Mridul, M. A., Jannat, M. E., & Islam, M. S. (2018, September). A Deep CNN Model for Student
Learning Pedagogy Detection Data Collection Using OCR. In 2018 International Conference on Bangla Speech and
Language Processing (ICBSLP) (pp. 1-6). IEEE.
Ahmed, N., Aziz, S. T., Mojumder, M. A. N., & Mridul, M. A. (2023, March). Automatic Classification of Meter in
Bangla Poems: A Machine Learning Approach. In 2023 6th International Conference on Information Systems and
Computer Networks (ISCON) (pp. 1-5). IEEE.
Contact
Sagar Sahu
CRAFT Research Project: Explainable Risk and Coverage with Computable Insurance Contracts
Educational Background
B.S. in Computer Science, Rensselaer Polytechnic Institute (Aug 2022 - Dec 2025)
Bio
Sagar is a recent graduate from Rensselaer Polytechnic Institute. Having studied Computer Science (with a concentration in AI, ML, and Data Science) as well as minored in Economics of Banking & Finance, his interests align with the intersection of artificial intelligence and modern financial services/systems. Throughout his academic career, Sagar focused on exploring cutting-edgetechnologies and applying them to solve real-world issues. He worked on a CRAFT project in the fall of 2025 and gained an astute understanding of fintech and risk analysis through bi-directional traceability mechanisms for both natural language and propositional logic. His breadth of exposure spans end-to-end development using NLP, computer vision, neural networks, and more. Sagar is grateful for his experiences and eager to use the knowledge he obtained to advance his future career endeavors.
Research Interests
Machine Learning
Blockchain Technology
Software Development
Cryptocurrency and Stocks
Money Markets
Agentic AI Systems
Research Projects
Developing Training Frameworks for Artificial Neural Networks
Finite Element Analysis using CNN Classification
Contact
Zahra Shoorvazi
CRAFT Research Project: AI Compliance Officer
Educational Background
M.S. in Financial Engineering, Stevens Institute of Technology, 2025
Bio
Emma is a quantitative researcher working at the intersection of artificial intelligence and financial systems. She completed her M.S. in Financial Engineering at Stevens Institute of Technology, where she joined the CRAFT research group under Professors Zachary Feinstein, Ioan Florescu, and William Long.
Her current research focuses on applying machine learning and large language models to financial regulation and market data. Emma’s parallel work includes earnings forecasting, volatility modeling, and data-driven financial prediction using modern machine learning techniques. More broadly, her interests lie in integrating AI/ML, knowledge representation, and quantitative finance to build reliable and interpretable intelligent systems for financial institutions and regulators.
Research Interests
AI in Financial Markets
LLMS
Knowledge Graphs
Multi-agent LLM Systems
AI-Driven Decision Support
Volatility Modeling
Stochastic Modelling
Optimization Methods
Research Projects
CRAFT Project: Knowledge Graph + LLM-Based Question Answering for U.S. Financial Regulation. Dec. 2024 to Dec. 2025
Analyst Ranking and EPS Estimation with ML/LLMs using IBES, Bloomberg, Compustat data (2024-2025)
Volatility Estimation for High-Frequency Portfolios using Wavelet Neural Networks (LLWNN) (2025)
Contact
Dominick Varano
CRAFT Research Project: Federated Learning for Fairness-Aware and Privacy-Preserving Financial Risk Assessment
Educational Background
B.S. in Computer Science at Stevens Institute of Technology
Bio
Dominick Varano is a fourth-year undergraduate student in Computer Science at Stevens Institute of Technology. He has a strong passion for Machine Learning and Artificial Intelligence. His research has focused on studying and developing individual and group fairness algorithms within federated learning frameworks.
Research Interests
Machine Learning
Artificial Intelligence
Natural Language Processing
Contact
Bolun “Namir” Xia
CRAFT Research Project: Predictive Learning from Long Financial Documents
Educational Background
Ph.D. Candidate in Computer Science, Rensselaer Polytechnic Institute (August 2022 - )
M.S. in Computer Science, Rensselaer Polytechnic Institute (August 2020 - May 2022)
B.S. in Finance and Computer Science, New York University, Stern School of Business (Fall 2016 - May 2020)
Bio
Bolun "Namir" Xia is a Ph.D. candidate in Computer Science at Rensselaer Polytechnic Institute, with an undergraduate background in Finance and Computer Science. His research focuses on Natural Language Processing (NLP) and its real-world applications in Finance. Being both versed in Finance and Computer Science, he aims to bridge the gap between both, contributing to the growing field of Financial Technologies in the NLP sphere. His work mainly focuses on developing better representations of textual data for predictive analytics, with cutting-edge NLP methods such as Transformer Language Models and Graph Deep Learning. Through his involvement with CRAFT, he has been able to produce favorable results in long document regression tasks, which predict numerical target variables using long documents, such as using 10-K reports to predict firm performance indicators and analyzing earnings calls to forecast its immediate impact in temporal stock price dynamics. He now aims to improve the craft by designing novel graph-based deep learning methods specialized for prediction tasks based on financial text. In his free time, he likes to practice traditional archery and learn different languages.
Research Interests
Natural Language Processing
Predictive Analytics
Long Document Regression
Graph Deep Learning
Machine Learning
Transformer Language Models
Publications
Bolun "Namir" Xia, Vipula D. Rawte, Mohammed J. Zaki, and Aparna Gupta. FETILDA: an effective framework for fin-tuned embeddings for long financial text documents. arXiv Computing Research Repository, Jun 2022.
Research projects
Predictive Learning from Long Financial Documents
Contact
Zhiyuan Yao
CRAFT Research Project: Extending, Simulating and Scaling Decentralized Exchanges Made by Automated Market Makers
Educational Background
Ph.D. in Financial Engineering, Stevens Institute of Technology (August 2019 - September 2024)
M.S. in Financial Engineering, Stevens Institute of Technology (August 2017 - May 2019)
B.S. in Mathematics, Nankai University (August 2013 - June 2017)
Bio
Zhiyuan Yao's research focuses on the intersection of reinforcement learning, machine learning, and financial engineering. He has earned a PhD in Financial Engineering, is currently working as a Quantitative Developer at Millennium. He specializes in developing innovative solutions for challenges in algorithmic trading, market simulation, and financial optimization. His work addresses critical issues such as performance degradation in real-world trading strategies, efficient handling of large-scale action spaces, and simulating realistic financial markets.
Zhiyuan's recent academic contributions include a hierarchical intraday trading framework that optimizes trading decisions end-to-end, reducing trading costs and improving profitability. He has also developed a market simulator driven by reinforcement learning agents, capable of replicating core real-world market characteristics, offering valuable insights into market microstructure and the impact of regulatory changes. Additionally, his research on model-based reinforcement learning with delayed observations has advanced decision-making processes in stochastic environments.
Research Interests
Reinforcement Learning
Market Simulation
Algorithmic Trading
Large Language Models
Publications
Z. Yao, et al. "Towards Modern Card Games with Large-Scale Action Spaces Through Action Representation." IEEE Conference on Games (CoG) 2022.
Z. Yao, et al. "Control in Stochastic Environment with Delays: A Model-based Reinforcement Learning Approach." Accepted in The 34th International Conference on Automated Planning and Scheduling (ICAPS).
Z. Yao, et al. "Reinforcement Learning in Agent-Based Market Simulation: Unveiling Realistic Stylized Facts and Behavior." Accepted in International Joint Conference on Neural Networks (IJCNN) 2024.
Q. Xie, et al. "The FinBen: A Holistic Financial Benchmark for Large Language Models." Accepted in Conference on Neural Information Processing Systems (NeurIPS) 2024.
Y. Yu, Z. Yao, et al. "FinCon: A Synthesized LLM Multi-Agent System for Enhanced Financial Decision Making." Accepted in Conference on Neural Information Processing Systems (NeurIPS) 2024.
Y. Cao, Z. Yao, Z. Chen, Z. Deng. "CatMemo@IJCAI 2024 FinLLM Challenge: Fine-Tuning Large Language Models using Data Fusion in Financial Applications." Proceedings of the Eighth Financial Technology and Natural Language Processing and the 1st Agent AI for Scenario Planning.
Research projects
Predictive Learning from Long Financial Documents
Contact
Yixuan Zeng
CRAFT Research Project: The Role of Wholesale CBDCs in Financial Stability: Models, Risks, and Global Case Studies
Educational Background
B.S. in Mathmatics dual with Economics at Rensselaer Polytechnic Institute (RPI), class of 2025
Bio
Yixuan Zeng is a senior at Rensselaer Polytechnic Institute, pursuing a dual major in Mathematics and Economics. With a strong academic foundation, Yixuan has excelled in courses such as Data Mathematics, Advanced Data Analysis in Economics, and Optimization These experiences have equipped her with essential skills in statistical modeling, machine learning, and data visualization。
Further, her research projects, including an investigation into the impact of wholesale CBDC on financial stability and the improvement of cost functions for deep learning models, demonstrate her ability to integrate academic knowledge with practical applications.
Beyond academics, Yixuan has gained hands-on experience in both technical and leadership roles. As a researcher, she contributed to machine learning and optimization projects. Her work has been recognized for its innovative approaches and collaborative impact.
Yixuan aims to bridge the gap between technical and strategic roles in the technology industry. Her long-term goal is to contribute to AI-driven product management and consulting, focusing on using data-driven insights to create solutions that meet diverse stakeholder needs.
Research Interests
Risk Analytics
Data-Driven Policy Making
Financial Innovation
Healthcare Analytics
Publications
The Role of Wholesale CBDCs in Financial Stability: Models, Risks, and Global Case Studies. (In Progress)
History of Research Projects
Improving Cost Function For IDLG
Biodegradability Prediction
Fairness Evaluation of LLM