2025 Code for GovTech DMP program
- David Higgins
- Rajan Maurya
This year in Code for GovTech DMP 2025 we are proposing 6 projects focussed around our Mifos Projects.
All applicants should review our Summer Intern video at :
Applications open on April 7th but check out the process now at:
Code For Govtech - DMP2025
The projects are as follows:
Mobile Applications:
Mifos Pay v4.0
Description: Mifos Pay is an open-source mobile wallet and digital payment solution designed by the Mifos Initiative—a global project committed to financial inclusion. It empowers underserved, unbanked, and underbanked populations, especially in emerging markets, by providing secure, accessible, and efficient digital financial transactions. Its primary goal is to bridge financial gaps, promote economic empowerment, and simplify digital transactions for communities traditionally left out of formal banking systems.
Goals:
Generic KMP Module for Merchant Payments: Develop a flexible Kotlin Multiplatform module allowing seamless integration with any merchant SDK for payments, enabling easy plug-and-play integration for diverse merchant services.
KMP Sandbox for Generic Loan Applications: Create a generic, extensible KMP sandbox environment enabling loan applications from any lending company, ensuring easy integration with multiple lenders through a standard interface.
Country-specific Payment Provider Integration: Implement a flexible payment-provider integration system that dynamically connects with valid local payment gateways based on the user's country, ensuring localized payment support.
Auto-Pay Bills Module: Implement a generic KMP module supporting auto-payment of bills by integrating configurable scheduling logic, payment-provider selection, and secure transaction execution, enabling users to effortlessly automate recurring payments.
Tech Skills Needed: Kotlin Multiplatform, GitHub Actions, Kotlin, Android, IOS
Expected Outcome: Target release for v5.0
Mentors: @Rajan Maurya
Mifos Mobile - End User Application v8.0
Description: An open-source mobile banking application powered by the Mifos Initiative, designed to enhance financial inclusion by providing unbanked and underbanked populations seamless access to banking services. It offers account management, loan applications and repayments, savings and deposits management, fund transfers, and transaction tracking through integration with the Mifos X backend via RESTful APIs. Targeted primarily at emerging markets, Mifos-Mobile empowers users to conveniently manage finances, even in remote areas, while actively supported by a global community dedicated to extending and improving digital financial services.
Goals:
Scheduled Payments: Allow users to schedule recurring payments and transfers.
Transaction Categorization: Automatic categorization of expenses for better financial tracking.
Redesign Dashboard with given Figma designs and Customizable Dashboard: Let users arrange and prioritize frequently used features.
In-App Customer Support: Real-time chat and customer support for immediate user assistance.
Release on IOS, Desktop platform
Tech Skills Needed: Kotlin Multiplatform, GitHub Actions, Kotlin, Android, IOS
Expected Outcome: Target release for v80
Mentors: @Rajan Maurya
Mifos Gazelle:
Mifos Gazelle Runtime GUI v1.0
Description: This project focuses specifically on a run-time GUI that would be able to narrate a demo use case to end users from deployment through to explaining how data is transferred between each of the deployed components throughout the demo. Linking to the existing UI’s of components to show these flows. Whilst this may focus on one specific demo journey for this project it should be written with sufficient documentation and flexibility as to how future demos can be incorporated in the UI. This is seen to be distinct but closely correlated to the Mifos Gazelle Profile/Demo Creater project. This GUI should use the existing commands/API’s and documentation within Mifos Gazelle.
Goals:
Define the Figma flows for the GUI
Proposal on GUI Tech agreed
GUI implemented
Demo Flow incorporated
Documentation of how future demos can be incorporated
Released in v 1.3.0
Tech Skills needed: We are not prescribing the front end technology but existing Mifos products use React and Angular. Where possible it should be reusing existing front end assets of Mifos.
Expected Outcome: Target release for publication is v1.3.0 of Mifos Gazelle (Sept/Oct 25).
Mentors: @David Higgins @tom daly
Mifos Gazelle OpenCRVS integration
Description: This project looks specifically at how OpenCRVS could be integrated within Mifos Gazelle to support end to end use cases including OpenCRVS alongside other Mifos Gazelle deployed DPGs. Mifos Gazelle supports 2 methods of integration:
(1) Integrated Deployment in Kubernetes.
(2) Integration via hooks as an external component (e.g. maybe pre-deployed or cloud hosted) to enable end to end use cases.
Goals:
The first step of the project would be to agree the ideal method of integration.
Before then undertaking the integration, with full documentation and postman collections to enable demos.
Proof of Integration would be by delivery of at least 1 demo use case incorporating OpenCRVS with at least 1 other Mifos Gazelle deployed component.
Tech Skills needed: Technologies used in Mifos Gazelle are Bash, Postman, Helm, Kubernetes, Docker.
Expected Outcome: Target release for publication is v1.3.0 of Mifos Gazelle (Sept/Oct 25).
Mentors: @David Higgins @tom daly
AI:
Accessibility and Language improvements to increase inclusion in Applications
Description: This project looks to increase inclusion and accessibility through improved language translations. Mifos applications have a number of UI’s and some existing translations. However feedback has been that even in current translation options, sometimes the translation has not taken into account the context. This increases barriers to adoptions.
This project could take a number of directions depending on the interns interest and decision process. For example do we use generative AI to help with the translation, is there an ML aspect to gather feedback and improve the translation. Is the translation done at the point of release and packaging of an application (mobile or web) or its it live translation. Do we focus on the buttons and menus, the help files or even look at language translation capabilities like a co-pilot for our developer documentation so we can include a wider community of developers. What opportunities are there to link in with the open source initiatives that focus on translations. This project is ideal for an Intern with an inquisitive mind, keen to explore options and happy to refine based on initial investigation. The reward for being confident to define a solution is a potentially huge impact on inclusion.
Goals:
Define an initial scope and focus for the translation for phase 1.
Investigate and decide on an approach that is extendable to other needs in the future.
Mock up and prove the solution works working with our developer and application teams
Generate the solution and appropriate documentation (including for extending it in the future)
release either as a release manager tool, developer tool, or within our apps.
Tech Skills Needed: ChatGPT, LLM, Generative AI, Kotlin
Expected Outcome: A first release of the improved translation capability either incorporated into an App (Mobile/Web release) or as a community tool for use by developers/release managers to create the translation files, or as a community tool for developers depending on the chosen focus.
Mentors: @David Higgins @Ed Cable @Parth Kaushal
Voice-Driven Banking via Large Acoustic Models (LAMs)
Description: The "Voice-Driven Banking via Large Acoustic Models (LAMs)" project develops a voice-based banking platform to enable financial transactions in low-resource languages and dialects. The system will support tasks like balance inquiries, fund transfers, credit applications, and more using voice commands, removing barriers of literacy and technical skills. This initiative targets rural and underserved populations, promoting financial inclusion through innovative AI solutions.
The idea behind this project is to use a LLM to give command/prompts and have those commands fulfilled via selenium
Goals:
Technical: Build and deploy LAMs for regional languages and accents, integrated with NLP for intent recognition and voice biometrics for secure authentication.
Business: Expand access to banking services, improve customer engagement, and reduce service costs via automation.
Social: Empower non-literate users in underserved areas with seamless access to financial tools.
Business: Increased customer base, lower operational costs, and improved satisfaction.
Social: Enhanced financial access for marginalized groups and improved digital literacy.
Tech Skills Needed: LLMs, Selenium
Expected Outcome:
Application: Mobile/web app supporting voice-driven banking in multiple regional languages.
Capabilities: Voice-initiated transactions (balance checks, payments, loans) with real-time processing.
Security: Voice biometrics and encrypted transactions for data integrity.
Analytics: Dashboards to monitor usage and improve language support.
The final product will simplify banking for rural users, bridging the digital divide and boosting financial inclusion effectively
Mentors: @Akshat sharma @Lalit Mohan S @Parth Kaushal