IS484 IS Project Experience (FinTech)

Course Description:

IS484 is cancelled for AY2021/22 Term 2. Will resume in AY2022/23 Term 1.
Please seek sponsors and plan for IS483 for AY2021/22 Term 2.

• This is an SMU-X course designed in collaboration with participating Banks, FinTechs, and other FIs, to serve as project sponsors. Collectively, industry sponsors will supply a minimum of 5 projects ideas to select from.
• Students will form teams of 5 or 6, and select one the project ideas to work on. Project selections do not need to be unique, meaning multiple teams can select the same project idea.
• Each student project team will be assigned to a sponsor/mentor and an SMU faculty supervisor.
• Sponsors will provide project scope and management for student teams to have practical industry learning experiences.
• Student teams will have weekly check in meetings, either virtually or physically, with their sponsor.
• Sponsors will specify the technologies to be used, including; development tools/languages, OS, database, 3rd party libraries, target deployment environment e.g. cloud environment.
• Student project teams will be expected to develop a working software application prototype, to be delivered to the sponsor at the end of the course.

Project Timeline:

Activities	Timeline	Term 1/ Term 2	Action By
Project Sourcing and Registration	Week -14 to Week -10	Form teams. Review the below set of predefined projects provided by Citibank, OCBC, NETS, UBS, and others. Fill up the Project Team Signup Sheet at the below link, listing your preferred projects. FT Track Coordinator will finalize the matching of teams to projects.	Students
Project Matching	Week -10	FT Track Coordinator will finalize the matching of teams to projects.	FT Track Coordinator
Proposal	Due before the start of Week -8	Submit your project proposals to your Track Coordinator(s). For mixed-track teams, both track coordinators need to review your proposal.	Students
Decision on Proposal	Week -4	Your Track Coordinator(s) will confirm that the project has sufficient scope to fulfill your respective track requirements for IS Project Experience.	Track Coordinator, Students, (Optional: Sponsor)
Start of Project	Week 1	Supervisor - Teams	Student
Midterm	Week 8	Presentation	Students, Supervisor, Reviewer (Optional: Sponsor, Track Coordinator)
Finals	Week 14 to Week 16	Presentation	Students, Supervisor, Reviewer (Optional: Sponsor, Track Coordinator)

IS484 Project Wiki:

Project teams to maintain their documentation here:
IS484 Project Wiki Home Page

Project Team Signup Sheet:

AY2020/21 Term 1
https://docs.google.com/spreadsheets/d/1IDAhC4JiK3RuKnIDQMG5UjJ6I1IiImo81Lu13wAuUxE/edit?usp=sharing
AY2020/21 Term 2 - CANCELED
https://docs.google.com/spreadsheets/d/1IDAhC4JiK3RuKnIDQMG5UjJ6I1IiImo81Lu13wAuUxE/edit#gid=1043528005 - CANCELED
AY2021/22 Term 1
https://docs.google.com/spreadsheets/d/1IDAhC4JiK3RuKnIDQMG5UjJ6I1IiImo81Lu13wAuUxE/edit#gid=86226209
AY2021/22 Term 2 - CANCELED
AY2022/23 Term 1
https://docs.google.com/spreadsheets/d/1IDAhC4JiK3RuKnIDQMG5UjJ6I1IiImo81Lu13wAuUxE/edit#gid=155160571

Current Projects

Item	Project Sponsor	Project Description	Project Deliverables	Project Stakeholders
1	NETS	Transform Internet Online Direct Debit via Web3.0 - Web 3.0 blockchain identity which offers privacy, control, openness and interoperability is a powerful catalyst to transform the usability and technology for Internet Online Direct Debit. The Web3.0 blockchain identity is a viable secure substitute for the end-user’s internet banking ID credentials and will also streamline the user experience for Internet Online Direct Debit from a Web2 to a Web3 experience. Students are asked to create a prototype which can accomplish the following: Demonstrate the registration of a Web3.0 id (blockchain id) thru the existing banking credentials via a prototype banking app. Demonstrate Web3.0 login and notification for internet direct debit payments. Demonstrate capability on use of blockchain host to substitute and support the online internet direct debit payment processing functionality which may include a blockchain smart contract to bridge Web3.0 and legacy payment processing of the banking institutions.	Students will be tasked to build a prototype which contains the following: A prototype bank app to simulate the registration of a Web3.0 id (blockchain id) thru the existing banking credentials and to be the mobile app to enable a Web3.0 login and notification for internet direct debit payments. A blockchain host to support the Web3.0 login and payment processing functionality which may include a blockchain smart contract to bridge Web3.0 and legacy payment processing of the banking institutions. A banking payment simulator host to register and demonstrate successful payment processing and reference to Web3.0 audit trail. Project sponsors will share sufficient context so students can understand the context of internet online direct debit to enable the students to design suitable solutions to overcome the problem statement. A sample high level conceptual design, explanation of this roles and responsibilities of the components and other useful details will be available.	Project Coordinator: David Woo Chee Keong DavidWoo@nets.com.sg Project Mentor: TBD email TBD
2	NETS	Interoperable QR payments using EMVCo QR - EMVCo QR implementations currently requires consumers to download and deposit funds in multiple payment apps in order to pay to the whole spectrum of EMVCo QR merchants as opposed to only needing to use their favorite payment app. EMVCo QR merchants need to sign-up, settle and reconcile with multiple payment providers as opposed to a single party. EMVCo QR labels currently need to be replaced physically when a merchant decides to ADD or REMOVE QR payment options. Students are asked to create an interoperable payment solution using EMVCo QR which can accomplish the following: Allow users to use their favorite payment apps to pay to all EMVCo QR merchants as opposed to users having to download multiple payment apps in order to pay to merchants accepting different payment apps. Allow merchants to accept payment from all payment apps and receive settlement including consolidated reporting from only one acquirer versus settlement with multiple acquirers each representing different payment app providers. Allowing a unified and streamlined QR payload which is merchant centric and do not need replacement if the merchant decide to switch acquiring relationships.	Students are expected to deliver a prototype as follows: A working eco-system prototype comprising mock\up payment apps, merchants, EMVCo QR labels, EMVCo QR switch (if applicable) and sample transaction and settlement reporting dashboards received by both consumers, merchants and the scheme. The prototype should be able to highlight and demonstrate key concepts which are key to enabling such an interoperable QR payment scheme. Project sponsors will share sufficient context so students can understand EMVCo QR and the QR payment landscape to enable the students to design suitable solutions to overcome the problem statements. A sample high level conceptual design, explanation of this roles and responsibilities of the components and other useful details will be available.	Project Coordinator: David Woo Chee Keong DavidWoo@nets.com.sg Project Mentor: TBD email TBD
3	NETS	Car Park As-A-Service - Parking can be costly especially for car users/owners who need to hop across multiple car parks for work across multiple locations within to carry out their daily business or work activities. Current season parking arrangements are also inflexible and requires the car users/owners to commit a recurring fee to only one car park location. During peak hours it can also be extremely time consuming for car users/owners to queue and seek for available car park lots. Carpark as a service is a new service concept which allows a car park operator to rent physical car park spaces from real-estate property owners and to offer use of these car park spaces to subscribers of this service for a one-time or recurring fee. This allows a car park operator to provide car park services across multiple private car parks across the city without the need to own or rent a complete physical car park. This also allows car users/owners to subscribe to a contemporary digital car park service which allows them flexible membership and subscription options which can apply across all available car parks offered by the service including pre-booking, notification of availability etc.	Students are expected to deliver a prototype as follows: A prototype consumer mobile app that provides a real time view of available car park as-as-service functions, subscription and booking options. A central car park simulator with allows business model simulation leveraging machine learning/deep learning algorithms to test a variety of simulated statistics of car users/owners, preference and requirements. A car-park-as-as-service business viability assessment report using Singapore data. Project sponsors will share sufficient context so students can understand the context of the car park payment service to enable the students to design suitable solutions to overcome the problem statements. A sample high level conceptual design, explanation of this roles and responsibilities of the components and other useful details will be available. Project sponsors will provide a baseline car park ecosystem simulator which allows the students to further enhance and develop to incorporate the required use cases and data model to support the business model simulation and machine / deep learning.	Project Coordinator: David Woo Chee Keong DavidWoo@nets.com.sg Project Mentor: TBD email TBD
4	Vertex Holdings	AI Augmented Analysis of Financial Statements - Processing and analyzing management reports and financial statements of start-ups, existing portfolio companies and publicly listed organizations are a critical part of making investments. These reports however are typically in semi-structured data formats, requiring human intelligence to extract and process relevant content. This project seeks to explore the possibility of leveraging machine learning and artificial intelligence to automate the extraction, processing and interpretation of structured and unstructured content from business and financial reports to augment the investment analyst.	Students are expected to deliver a prototype as follows: Identify and implement tools to read and extract structured and unstructured content from business and financial reports in different formats e.g., PDF, PPTX, Excel. Interface for reports to be uploaded by organization, and to be able to search, interact and visualize data within and across organizations. NLP on textual content to include entity extraction, sentiment, topics, summarization. Project sponsors will share sufficient context so students can understand how/where this UI brings value to users. Project team will provide samples of business and financial reports for processing.	Project Coordinator: Joey Chua joey.chua@vertexholdings.com Project Mentor: TBD email TBD
5	Vertex Holdings	Company Smart News Monitoring - Investment managers need to keep abreast on latest news and developments relevant to companies they have invested in order to stay on top of their game. Current news monitoring is typically company keyword based e.g., "Vertex Holdings" which only surfaces news of direct relevance to the organization. However, various news events such as those regarding competitors, industry relevant information are important for managers to be aware about. This project will involve designing and developing a module for investment managers to review, track, and read key news items that affects the companies they have or plan to invest in.	Students are expected to deliver a prototype as follows: Develop news ingestion pipelines for searching and ingesting content based on keywords of interest. Design algorithms or heuristics to identify topics and keywords for news content that are indirectly related to a company of interest. Backend ranking methodology to score relevancy of specific articles to the investment manager. Build web application and user interface that provides an interactive way for the analyst to review, interact and make sense of the content highlighted to him. Project sponsors will share sufficient context so students can understand how/where this UI brings value to users.	Project Coordinator: Joey Chua joey.chua@vertexholdings.com Project Mentor: TBD email TBD
6	Vertex Holdings	Robo-advisory for Go-To-Market - Investment Managers not only invest in companies, but also help with the companies’ business development in different countries. However, this not only requires significant time to search for vendors/customers, but also require local knowledge of each country e.g., how businesses operates. Currently, this requires a lot of manual effort and individual knowledge to put together. We envisage an AI augmentation tool to support this task.	Students are expected to deliver a prototype as follows: Acquisition of relevant data sources e.g., customers and vendors to related portfolio companies. Collection of datasets and information to understand various industries and markets. Web application for user to input company and market fields as well as additional supporting information required. Website interface should allow user to generate report on demand e.g., export to PDF. Generation of potential leads as well as useful information sources and opportunities for entering the market as well as contacts or events to reach out to. Project sponsors will share sufficient context so students can understand how/where this UI brings value to users. Provide list of portfolio companies and the industry the companies are in.	Project Coordinator: Joey Chua joey.chua@vertexholdings.com Project Mentor: TBD email TBD
7	SMU / Narwhal Financial Systems	Automated Loan Origination System - The loan origination process for traditional banks is typically long running with many process steps involving human interaction, however for digital banks these processes are automated. This project is to prototype an automated end-to-end process for retail banking loan origination, for example a mortgage loan. The Loan Origination System (LOS) is to be deployed as an SMU tBank application, which will fulfil two purposes: a) to provide hands-on lab experience in the classroom, and b) to serve as a prototype for Narwhal Financial Systems (a.k.a. Narfin) which is an SMU spin-off company. The LOS will implement a completely automated process, starting from a customer online loan application, and ending with a loan approval (or rejection), without any “bank staff” involvement.	Students are expected to deliver a prototype as follows: Extraction of document fields using OCR technology, to complete the electronic record of the loan application. Computing a credit score, using a credit scoring model (e.g., scorecardpy). Note: the credit scoring model should be exposed as a reusable API. Implement credit decisioning rules based on the computed credit score, including pricing of the loan (i.e., the offered interest rate based on the customers risk profile). User Interfaces: Customer UI – for; a) applying for the loan online and uploading documents, b) checking loan application status, and c) accepting or rejecting loan offers. Bank Staff UI – for; a) manually making credit decisions on borderline loan applications, and b) viewing the status of all loan applications using various filters. Admin UI – for; a) configuring the credit scoring model, e.g., setting limits and constraints, and b) training the model using third party datasets, e.g., from lending club.	Project Coordinator: Alan Megargel alanmegargel@smu.edu.sg Project Mentor: TBD email TBD
8	SMU / Narwhal Financial Systems	Stock Trading Engine - A stock order management system and stock trading engine are to be deployed as an SMU tBank application, which will fulfil two purposes: a) to provide hands-on lab experience in the classroom, and b) to serve as a prototype for Narwhal Financial Systems (a.k.a. Narfin) which is an SMU spin-off company. The main features of the prototype are to include the following: Synchronization of stock quotes form Yahoo Finance. State Machine to facilitate the easy management of all transitions between trade states. Trade Timer to trigger the expiry of associated trades (e.g., limit orders). Trade Matching Engine to match counter party trades conducted based on the order type, price, and pending volume of each trade. Trading Algorithm whereby the trader can set parameters and execute automatic trading.	Students are expected to deliver a prototype as follows: Account Management: CRUD of Accounts. Stock Management: CRUD of Stock, Synchronization with Yahoo Finance. Trade Management: CRUD of Trade, State machine, Trade timer. Trade Matching Engine: Trade matching algorithm, Logging of all activities. Trader Book Management: CRUD to the book, Portfolio analysis. Trading Algorithm: Parameters exposed to trader, Execution of auto-trading. Gamification / Leaderboard for classroom usage, to reveal the top traders for a lab session.	Project Coordinator: Alan Megargel alanmegargel@smu.edu.sg Project Mentor: TBD email TBD
9	UBS	Client Experience Data Mesh & Dashboard - Data Mesh is a strategic approach to modern data management. The main objective of Data Mesh is to evolve beyond the traditional centralized data management methods of utilizing data warehouses and data lakes. Data Mesh emphasizes on the idea of organizational agility by empowering data producers and data consumers with the accessibility to access and manage data, without the trouble of delegating to the data lake or data warehouse team. The decentralized method of Data Mesh allocates data ownership to domain-specific groups that serve, own, and manage data as a product. We plan to build a data mesh to collect and manage data related to client experience.	Students are expected to deliver a prototype as follows: Identify the key data elements that can be collected via digital channels that can be managed as a Data Mesh. Create the domains and architecture to manage this mesh. Define and build Data products that can be consumed for measurements and calculating scores. Create a dashboard of client experience scores such as: Customer Effort Score, Net Promoter Score, Customer Satisfaction Score, Task Success Rate, Error Occurrence Rate, System Usability Scale, Time-on-task. Project sponsors will share sufficient context so students can understand how/where this brings value to further enhance client experience and design.	Project Coordinator: Saju Philip saju.philip@ubs.com Project Mentor: Sweata D , Kai Xian email TBD
10	UBS	UBS Project 2 - Coming Soon.	Students are expected to deliver a prototype as follows: Coming soon	Project Coordinator: Saju Philip saju.philip@ubs.com Project Mentor: TBD email TBD
11	OCBC	OCBC Project 1 - Coming Soon.	Students are expected to deliver a prototype as follows: Coming soon	Project Coordinator: Dedy Lim limdd@ocbc.com Project Mentor: TBD email TBD

Archived Projects (no longer available)

Item	Project Sponsor	Project Description	Project Deliverables	Project Stakeholders
X	Citibank	Derivative & Structured Product Performance Dashboard - Derivative & Structured products are complex and its crucial for Bankers and investment counselors to have a consistent view for how these products perform for our clients. Apart from product performance it’s important to know product lifecycle events and any risks that may be detrimental to private bank clients. This dashboard will allow visualization of such complex information in an organized and intuitive manner. Bankers and Investment counselors (ICs) act on market trends and guidance from research teams to create customized financial products for clients. These products are created to cater to a variety financial risks and client preferences.	The ask is to create an analytics dashboard that: Allows users to view cumulative financial performance of the products. Surface product performance details, including possible risks from changes in the market conditions etc. Filter and show a summary of upcoming product milestone details – such as interest payments, premiums due etc. Allow this data to be sorted and filtered to show details for one or more clients. Visualize this data using charts, tables etc. in a simple, uncluttered fashion.	Project Coordinator: Kulkarni, Kaushik kaushik.achala.kulkarni@citi.com Project Mentor: Awan, Kashif
X	Citibank	Preventive Cross-Platform Risk Assessment - Multiple applications are constructed together to support one of the largest Custodian banking platforms. Any of the components malfunctioning will affect productivity and also lead to a breach of the market deadline. We are seeking for an AI risk monitoring and assessment tool to enhance the platform resilience to another level. AI machine learning Platform to provide risk assessment of cross application health status and prediction of downtime. To do this, they need real time access of: Application through-put performance. End-to-end application cross-platform health assessment. Daily average volume vs. real time system load.	Students will be tasked to build a UI which: Contains a dashboard that provides a real time view of platform health status. Leverages machine learning / deep learning algorithms which suggests and predicts potential system downtime, potential SLA breaches, and identifies trigger points / bottle necks. Is able to construct end-to-end flows across different platforms.	Project Coordinator: Ho, Ricky ricky.ho@citi.com Project Mentor: Balusa, Ashok
X	Citibank	(Old name: Document Scrutiny using a Rules Engine) Document processing using Cognitive OCR - Currently the Document Scrutiny process is a manual task which requires human intervention for regulatory validations. This process is error prone and time consuming. A Rules Engine is need with these features: Perform Data Validations & Scrutiny for the received Transactions & Documents. Rules can be configured through UI & saved to the application at any point of time. A rich UI experience is needed for user friendly & easy rules configuration.	A solution or program which can accomplish the following: Download the Documents from Regulators portal for 5-6 countries for Consumer & Corporate banking platform. Decipher the Rules & Configure the Rules inside the Rule Engine. Receive the Transactions & the relevant supporting documents. Optical Character Recognition (OCR) & Named Entity Recognition (NER) will be performed by the system. Perform the Rule validations in an automated way for Transactions & Documents data extracted via the OCR Engine (Currently done manually).	Project Coordinator: Gupta, Arvind shweta4.gupta@citi.com Project Mentor: Mohammad, Thanveer
X	Citibank	Predictive Analysis of Risk Utilization - Phase II - Predictive Analysis of Risk Utilization enables Citi's clients and client facing officials to prevent regulatory violations, navigate trading disruptions by proactively take measures to prevent such breaches by allocating funds or by changing their trading strategy. Citi's institutional clients place millions of orders on any given trading day through its electronic execution platforms. As orders come in through Citi's systems, they are evaluated against several risk parameters (such as credit limits - Max Daily Notional, Daily Notional, Short Notional, etc) before the order is sent to the market. This project requires students to build capabilities to the system to predict and alert the clients of potential breach events both in isolation and combination of individual risk parameters.	Students executing this project will be expected to arrive at comparative machine learning solutions (Random Forest, LTSM and SVM) to predict imminent movement of the risk parameters based on historical trading patterns. Tasks include: Building adapters to funnel data to a central data pool to run analytics on the data. Analyzing and find inflection data points and patterns. Building a user interface/ data conduit that can be used by Citi clients/ users to be notified of any breaches if found.	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Kumar, Sudeep
X	Citibank	Equities Pre-Trade Booking Reconciliation - Equities Pre-Trade Booking is a manual task at present involving exchange dropcopy feeds, Citi’s internal trade feed for each client. The objective is to develop a tool where clients can review and confirm trades for a given product and market irrespective of execution brokers using exchange dropcopy and broker level reconciliation using blockchain which can be shared across brokers.	Equities Pre-Trade Booking Reconciliation using Blockchain Ethereum 2.0 Students to analyze the limitations and advantages of using Blockchain Ethereum 2.0 platform for financial data reconciliation. Develop UI to demonstrate the contents of 2 trade feeds at each block mutation. Give the final output at EOD in a file format with trade reconciliation exceptions.	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Kumar, Sudeep
X	Citibank	Machine Learning Model Performance - Machine learning models are being trained based on historical data. But in the commercial world, change is expected rapidly which may mark the model biased to the new data as well as scaled old data. Before the model is retained, there are immediate needs to understand what are the leverages that can be applied to interfere with the old model output to achieve the accuracy rate, then capture the business opportunity in a very short turnaround time. When models are unable to digest new data, they will generate inaccurate recommendations and predictions to the business, resulting in missing the opportunities for increased revenue.	A solution or program which can accomplish the following: Detect the root cause of low accuracy with a given model input, model output and model binary. Generate corrective recommendations to increase accuracy without re-building the model. Perform regression testing with recommendations, to demonstrate the expected accuracy. The program is expected to be able to analyse any supervisory learning model for the given input and output.	Yuqian Song, Head of APAC/EMEA Data Services and Head of Global Advanced Analytics Technology Solutions yuqian.song@citi.com
X	Citibank	Robo-Advisor - Student defined project. A robo-advisor that will; classify customers based on their investment experience and risk appetite, recommend a portfolio of investments to customers, provide visualizations / analysis of the customer's investment portfolio, and provide a budgeting and savings dashboard as an extension or the above.	A solution or program which can accomplish the following: Customer Classification (via chat) Portfolio Selection (recommendation to customer) Visualization (portfolio analysis) Personal Finance Dashboard (extension on top of the above)	Ravinder Rao, Senior Vice President, GCT Data & Analytics ravinder.rao@citi.com
X	Citibank	Private Banking Client Dashboard - Citi Private Bank (CPB) Investment Counsellors and Advisors provide frequent consultation to HNWI and UHWNI (high and ultra-high net-worth individuals) on how to manage their Investment portfolios. In order to perform their job they need high speed access to a client's positions, real-time market data and publicly available sentiment on the portfolio's constituents. The portfolio is usually composed of capital market securities and various funds (hedge, mutual, real estate, private equity). Careful thought needs to be put into providing an enriching UX / UI and leveraging machine / deep learning capability to provide robust recommendations. The users will use the information to pro-actively and also reactively service CPB's HNWI and UHNWI clients.	A working dashboard that provides a real-time view of a client's position. The view should be contextual based on the type of holdings (Cash/Liabilities, Equity, Fixed Income, Derivatives and Alternative Investments). The view would give an instrument and profitability analysis based on market data (Bloomberg / Reuters). Furthermore, there will be a recommendation engine that looks at a client's current / past positions and suggests trade-able ideas to the advisor based on upcoming announcements, trending public sentiment and client's personal interests.	Kashif Awan, Private Bank Capital Markets APAC Technology Head kashif.awan@citi.com
X	Citibank	Predictive Analysis of Risk Utilization - Citi's institutional clients place millions of orders on any given trading day through its electronic execution platforms. As orders come in through Citi's systems, they are evaluated against several risk parameters(such as credit limits) before the order is sent to the market. While currently, breaches in these parameters can be identified the moment the orders are placed, the next gen evolution of this risk management system requires predictive analytics of such breach events. This will enable Citi's clients and client facing officials to prevent regulatory violations, navigate trading disruptions by proactively take measures to prevent such breaches by allocating funds/ changing their trading strategy etc.	Students executing this project will be expected arrive at a machine learning solution to predict imminent movement of the risk parameters based on historical trading patterns. The solution should be able to take data feed for supplemental information (Triple witching dates, FTSE/MSCI rebalancing, other events that affect the market such as the Coronavirus threat) to more accurately predict exceptional scenarios. Tasks: Understand Citi's current data model for storing historical data. Build adapters to funnel data to a central data pool to run analytics on the data. Analyze and find inflection data points and patterns. Build supplemental data feed to establish market sentiments in the sytem and use that to augment their prediction models. Build a user interface/ data conduit that can be used by Citi clients/ users to be notified of any breaches if found.	Sudeep Kumar, Global Exchange Connectivity & Asia Cash Equities Technology Lead sudeep1.kumar@citi.com
X	Citibank	Customer Mailing Address Analysis - Addresses of people and businesses contain important information about them. More data about the locations of those addresses is required to get some insight from addresses. For example the population, geographic and economic indicators, crime rates etc. can be helpful. We need to collect such information about countries and cities to make the addresses usable in models and other analytics.	A solution or program which can accomplish the following: Collect information about countries from IMF data. Collect information about cities from DBPedia data. Build schedules to keep the above data fresh, as new data is available. Make this data available to lookup by country and Citi names to be used by models and analytics queries. Generate an embedding of countries and an embedding of cities, to be used as features in models. Unstructured addresses (where country, city are not marked separately, but part of large address text) need to be parsed before lookup. Make this information available by joining the addresses of people and businesses and collected data by countries and cities as join keys. Measure how much the model performance improves, after using this additional information.	Yuqian Song, Head of APAC/EMEA Data Services and Head of Global Advanced Analytics Technology Solutions yuqian.song@citi.com
X	Citibank	Marketing Audience Segmentation - Citibank sends merchants’ offers to the relevant customers. For example customers who often buy sports gear should get sports related offers and foodies should get offers from the restaurants. This requires accurate segmentation of customers as well as merchants. 3rd party data can be used to improve marketing audience segmentation.	A solution or program which can accomplish the following: Acquire 3rd party e.g. Statista, Euromonitor and map the brand mentions in the transactions, with brand master list in acquired data. Use brand category-hierarchy to segment the customers for their buying habits, using customer transaction history. Use brand category-hierarchy to segment merchants by categories of products and services sold and offers made. Use the category based segments for a broader match between customers and merchants.	Yuqian Song, Head of APAC/EMEA Data Services and Head of Global Advanced Analytics Technology Solutions yuqian.song@citi.com
X	OCBC	Online Business Account Maintenance - Business Banking provides SME & Corporate customers with a broad range of Cash & Trade products and services. Through our Digital Business Banking channels, customers are able to manage cash, loans, trade finance and perform transactions in their day-to-day business. The current account maintenance form is in PDF for customers to download. Once customer fills up the form, they will email the scan copy to operations. This process takes at least a few days to complete and incurs operation overhead. The task is to digitize the online business account maintenance services: To analyze the account maintenance form in PDF Develop UI & microservices to render the form based on configuration & capture the data digitally Allow customer to authorize the submission digitally with online signature or via SingPass	UI & microservices as a dynamic online account maintenance application. UI is preferred to be developed in ReactJS & microservices developed in Java Springboot Project sponsors will share sufficient context so students can understand how/where this UI brings value to the customers. The sample pdf form, explanation of the validations and other useful details will be shared.	Project Coordinator: Lim Dedy Daryono limdd@ocbc.com Project Mentor: Kotla Mutha Ravi Tej RaviTejKotla@ocbc.com
X	OCBC	Host-to-Host (H2H) Modernization - Business Banking provides SME & Corporates customer with a broad range of Cash & Trade products and services. Through our H2H integration, customers are able to integrate with their ERP/Accounting system to perform transactions with the bank. The current H2H solution is a secured file transfer channel which supports file transfers between customers and the bank over the internet. Currently there is excessive overhead in managing folder & access controls, as there are thousands of folders that need to be maintained. In addition, H2H integration is scheduled instead of event based, and this also incurs high maintenance overhead. The task is to modernize the H2H setup & maintenance: To analyze the H2H setup & challenges Propose a new design, implementation & tool if any to ease the maintenance with better access control	New H2H solutions setup for minimal operation maintenance Event trigger capabilities (instead of scheduled) H2H is preferred to be on Linux platform with Java components if any Project sponsors will share sufficient context so students can understand how/where the new H2H integration will benefit the customers.	Project Coordinator: Lim Dedy Daryono limdd@ocbc.com Project Mentor: Kochupurackal Joseph George JosephK@ocbc.com
X	OCBC	Blockchain Applications - GO&T serves as trusted partner to co-create new business capabilities, protect technology infrastructure, manage Group’s IT operations, and run regional processing hubs. Currently there is wide range of documents used by different teams in GO&T, there’s no efficient and automated method for verifying the latest version of documents, retrieval of documents and checking authenticity. This creates operational overhead for staff who rely on getting the approved documents done by other teams. On the NFT digital assets front, due to strong demand on NFT assets, there is no standardized way and repeatable process for production of NFT assets. Business users are not able to iterate fast on digital assets creation to meet demand by customers. The task is to design and develop blockchain applications for use cases such as: Document notarization & verification Production of NFT assets	New microservices for API calls to the endpoints of blockchain for document verification. New microservices for API calls to the endpoints of blockchain for production and management of NFT assets. New smart contract to store the Hash of the documents on-chain and other meta-data stored off-chain. New smart contract for NFT assets based on ERC-721 token standard stored on-chain and other NFT meta-data stored off-chain. Project sponsors will share sufficient context so students can understand how and where the blockchain applications can benefit the end users. The sample documents, explanation of the validations and other useful details will be shared.	Project Coordinator: Lim Dedy Daryono limdd@ocbc.com Project Mentor: Neo Wei Cheong WeiCheongNeo1@ocbc.com
X	Citibank	Commodities Pricing - The Commodities Pricing Platform (Atlas) is a Sales and Trader facing application that provides timely pricing across all Commodities asset classes. The user interface is mainly configuration driven and uses a highly dynamic rule engine to represent relationships between user inputs. These configurations are used as inputs to the rules engine to define products, their defaults and what data transformations should occur based on user inputs. The configuration files the pricing platform uses can span thousands of lines of JSON. Updating this JSON manually represents an operational risk. The aim of this project is to structure the update process and reduce this risk.	Students will be asked build a UI that will: Have the ability to create, delete and modify rules engine configurations Apply validation logic across the configuration to highlight inconsistencies in the rules such as loops, duplicates or conflicts. The application should provide an intuitive UI/UX, and the ability to create new pricing products without code changes. Project sponsors will share sufficient context so students can understand how/where this application brings value to users. The data will be shared and an explanation of this data structure and other useful details will be available.	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Ronnie Day email ronnie.day@citi.com
X	Citibank	Preventive Cross-Platform Risk Assessment (II) - An AI machine learning platform is needed to provide risk assessments of cross application health status and predictions of downtime, based on realtime access to applications through-put performance data, in order to provide an end-to-end cross platform health assessment including daily average volume vs realtime system load. Students are tasked to build a UI with a dashboard that provides a realtime view of platform health status, leveraging machine learning/ deep learning algorithms to suggest and predict potential system downtime, potential SLA breaches, and identify trigger points /bottle necks.	Students will be tasked to build a UI that will provide: Working dashboard with realtime view of platform health status. View of contextual assessment of platform status. View of trigger notifications when risk crosses threshold. View of collected historical information. Perform system end-to-end calculated risk assessment. Ability to collect data from different applications in realtime.	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Ricky Ho email ricky.ho@citi.com
X	Citibank	Equities Chatbot - Citi Equities trading platform receives several queries on a daily basis. The queries currently are being sent and answered through email, symphony chat, Bloomberg chat etc. and an associate’s manual efforts are required to generate a response; either running database queries to fetch results, or leveraging their product knowledge and judgement to make predictions. This project is to create an AI chatbot to automate the process and replicate the efforts of an associate replying to queries received by the Equities team. The AI chatbot is expected to interact with traders and answer their queries. The bot must be capable of: Running 24/7 to tend to all queries Generating cohesive and accurate responses to queries in real time Processing financial jargon and acronyms as input Running database queries and displaying results Making trading algorithm predictions for products selected by trader Logging all activity to be used for analysis of performance	Students will be tasked to build the following: Functional chatbot with clean UI and robust backend Chatbot must use NLP tasks such as Word sense Disambiguation, Named Entity Recognition (NEM) and Sentiment Analysis to ensure nuanced interactions Must mine text from existing datasets to enrich the bot Must employ decision making models to make trading algorithm predictions Project Inputs: Historic queries and responses Financial acronyms compilation Product to algorithm mapping Any FAQs	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Rajeshkumar Madanlal email rs97865@citi.com
X	Citibank	Data Aggregator - Reporting and Statements generation across various countries within Citi is a critical operation for client and service operations within Citi. There is a high level of data redundancy (multiple copies of the same data), manual dependency from operations users to input data, and untimely data availability. In some cases, this becomes a regulatory concern. Students executing this project will be expected to build a middleware application that is capable of extracting data points on demand from various source or upstream systems. The objective is to build a Report On Demand (ROD) framework that can serve as a self-service fully autonomous tool for operations users, to enable: Flexibility and enable users to retrieve data on-demand Centralized data aggregation management Cost-effective and timely response to ever-changing data requirements	Features of this solution to include: Ability to upload data a dictionary in a defined format – this format can be specified by the students, Ability to create data extraction templates from the available data points in the data dictionary. Once templates are created, they can be run on-demand to extract and store these data points from source systems. Note: The data sources typically are Oracle, MSSQL and No-SQL databases. This can be further extended to File store, HDFS etc. Project sponsor will share further context on the various pain points and priorities of the problem statement. Specific use will be defined and shared with the students too.	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Sandeep Sharma email rajeshkumar.madanlal.sharma@citi.com
X	Citibank	Enabling Efficient On-the-job Training with AI - Citi Commercial Bank (CCB) plans to hire hundreds of staff over the next three years to fast-track growth. It is crucial that new hires joining CCB are brought up to speed efficiently on company and role-specific processes and day-to-day functions. A large part of this is done through on-the-job training, where new hires learn more about the various processes, digital platforms and resources as they do their daily work. However, given the complex nature of the business (and hence its processes and systems), it is not always easy for new hires to learn on-the-job. The ask is to create on-the-job training resource(s) and platform for employees where Line Managers / L&D partners can easily: Build learning pathways for their teams on the various digital processes and systems that they would have to get familiar with as part of their on-the-job training. Upload / maintain use-case driven content for the learning pathways. Reduce manual on-the-job trainings for new hires.	Students are expected to deliver: A working prototype to collate and prioritizes training resource(s) that enables efficient and effective on-the-job learning. The solution should be easy and intuitive to use for both managers and new hires. The working prototype would demonstrate the use of natural language processing/AI and full stack development skills to build a platform that can deliver summarized and relevant information when users make specific on-the-job queries. The solution must be able to measure the success of on-the-job learning. New hires should easily be able to navigate the learning pathways such that they can self-serve and learn on their own and Search for answers to specific on-the-job queries they have on processes / systems. Project sponsor will share further context on the various pain points and priorities around onboarding and training new hires. Specific use case and target audience personas will be defined and shared with the students too.	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Go Cheng Yan email cg14129@citi.com
X	Citibank	Data APIs using Low Code - Lots of our data sourcing is driven from embedding SQL or Logic in Stored procedures. This makes the application logic heavily dependent on data which sometimes is not even owned by the given service. Accessing data in an easy way within a distributed platform is a big challenge. A low-code development platform (LCDP) provides a development environment used to create application software through a graphical user interface. A low-coded platform may produce entirely operational applications, or require additional coding for specific situations https://en.wikipedia.org/wiki/Low-code_development_platform The objective is to develop Data APIs using Low code technology automation. This is an agile way of software development. It will deliver easy access of our data (reference / product / transactional) to our Web, Mobile UI and reposting modules.	Students are expected to deliver: POC for Low Code products. Design new APIs based of selection made by project sponsor. Working APIs for data from current sources. Project learnings: Experience in designing / developing APIs. Learning an agile way of application development. Proof of connects with product(s) used of Low Code automation. Experience working on data complexity in a distributed system. Good idea about Capital Markets domain. Project sponsors will share sufficient context so students can understand how current data flow works and what we expect from the APIs, and details about Low code products which can be used by Citi.	Project Coordinator: Dossii, Shailej P shailej.p.dossii@citi.com Project Mentor: Rohit Rohatgi email rohit.rohatgi@citi.com