Profile Information
Name: Anisha Mahuli
GitHub: https://github.com/amahuli03
Resume:
Location: Delaware, United States
Typical working hours: 9:00 - 19:00 EST, but I'm flexible and can be reachable outside those hours. I can dedicate 40 hours a week to GSoC.
Synopsis
Short summary describing your project and how it will benefit Wikimedia projects
This project addresses a critical bottleneck in how the Wikimedia Foundation measures the success of global editing initiatives. Currently, system-wide data from the Programs and Events Dashboard must be manually extracted by maintainers. This manual process delays reporting and consumes valuable developer time. I will replace this with an automated, self-service data engine that allows staff to trigger and download comprehensive metrics on demand.
My approach focuses on building a high-performance backend using Ruby on Rails and Sidekiq. Because aggregating data across millions of edits can be resource-intensive, I will implement an asynchronous export system that processes heavy workloads in the background, ensuring the Dashboard remains responsive for all users. I also plan to overhaul the existing metrics suite to include deeper insights, such as long-term editor retention and global impact trends. By applying my experience in asynchronous processing and database optimization, I will provide the Wikimedia community with a stable, scalable tool to quantify their impact without the manual overhead.
Possible Mentor(s)
@Ragesoss, @Abishekdascs
- Have you contacted your mentors already?
I have reviewed the project documentation and am engaging with the community on Zulip and Phabricator as of today (March 16) to begin microtasks and refine this proposal.
Deliverables
Describe the timeline of your work with deadlines and milestones, broken down week by week. Make sure to include time you are planning to allocate for investigation, coding, deploying, testing and documentation
| Phase | Dates | Summary of Tasks |
| Community Bonding | May 4 – June 1 | Setup Docker. Become familiar with the app architecture and documentation. Create Phabricator task. Study Courses/Users schema. |
| Week 1-2 | June 1 – June 14 | Design CSV ExportService. Create API endpoint. Setup Sidekiq background jobs. |
| Week 3-4 | June 15 – June 28 | Build CSV logic. Optimize SQL queries. Fix N+1 issues. Submit 1st report. |
| Week 5 | June 29 – July 5 | Build Admin UI. Milestone: Functional system download. Complete evaluation. |
| Week 6-7 | July 6 – July 19 | Audit Stats page. Build new retention metrics. Submit 2nd report. |
| Week 8-9 | July 20 – Aug 2 | Integrate reference-counter API. Build Global Stats view. Add Redis caching. |
| Week 10 | Aug 3 – Aug 9 | Write RSpec tests. Perform load tests on CSV export. Fix stability bugs. |
| Week 11 | Aug 10 – Aug 16 | Staff feedback session. Refactor UI based on usage. Submit 3rd report. |
| Week 12 | Aug 17 – Aug 23 | Final documentation. Update README. Final submission and blog post. |
Participation
Describe how you plan to communicate progress and ask for help, where you plan to publish your source code, etc
I plan to communicate regularly with my mentors and maintain transparent progress throughout the project. If I hit blockers or design questions, I will ask for guidance when I need it rather than waiting for scheduled check-ins or remaining blocked. Most development discussion will happen through Phabricator, Zulip, and GitHub, since this is the project's standard method of communication.
All of my development work will be done in a fork of the WikiEduDashboard repository using feature branches. I will publish my code through GitHub pull requests and ensure that commits are clearly documented so that the reasoning behind changes is easy to understand. I will also follow the repository’s contributing guidelines and coding standards. To make progress visible, I will maintain a clear development plan with milestones aligned with the project timeline. Each milestone will correspond to a set of issues or pull requests so that mentors and community members can easily track the project’s progress. I will also make sure that any new features are well documented and tested before merging.
About Me
Tell us about a few:
Your education (completed or in progress):
I recently graduated from the University of Virginia with a bachelor's in Computer Science. I'm planning on continuing my education through a Master's Computer Science program starting this Fall 2026.
How did you hear about this program?
I have been following the Wikimedia community's work for some time and discovered GSoC through their contribution guides
Will you have any other time commitments, such as school work, another job, planned vacation, etc, during the duration of the program?
No, I have no other time commitments this summer. I will not be actively doing any jobs or academic programs, and I also have no trips planned.
We advise all candidates eligible for Google Summer of Code and Outreachy to apply for both programs. Are you planning to apply to both programs and, if so, with what organization(s)?
I'm only applying to GSoC.
Have you been accepted to GSoC before?
No, I have not.
Are you eligible for GSoC?
Yes.
Are you submitting any other proposals?
No.
What does making this project happen mean to you?
To me, this project is about fixing a major friction point for the people who actually keep Wikimedia running. Right now, if the Foundation staff needs to see how their programs are doing, they have to ask for data manually. That is a bottleneck that needs to be fixed. I love the challenge of taking a complex mess of data involving thousands of programs and millions of edits and turning it into something clear, accurate, and downloadable.
Coming from a background in Python and Go, I know that the hardest part of a project like this is the performance and reliability rather than the syntax. Aggregating system-wide metrics can easily become a resource hog, and building an export system that does not time out or crash the server is a puzzle I am eager to solve. I have handled similar challenges with background workers and API optimization in other languages, and I am looking forward to applying those same backend principles to the Ruby on Rails codebase.
Beyond the technical side, I am interested in the usability of the data. I want to make sure the metrics we build actually answer the questions the staff is asking, such as editor retention or the global reach of a specific campaign. Ultimately, I want to make sure the people organizing these events have the metrics they need to show the real-world impact of their work and make better decisions for the future of the community.
Past Experience
Please add links to any feature or bug fix you have written for a Wikimedia project during the application phase.
So far, I've made 2 PRs, both of which are still open as I'm waiting for a review. I'll keep this section updated as I continue contributing.
- bugfix: make the Contributions page restore scroll position after activity recreation
- feat: add "other" option as required field for image deletion nominations
I also reviewed a PR according to the guidelines.
Describe any relevant projects that you've worked on previously and what knowledge you gained from working on them.
My previous experience developing backend systems and API-driven applications has prepared me well for implementing system-wide metrics and data export functionality for the Programs & Events Dashboard.
During my software engineering internship at a cybersecurity startup, I worked on production backend services written in Go, where I developed and maintained REST APIs that supported cybersecurity products used by real customers. I collaborated closely with engineers, product managers, and security experts to design reliable backend endpoints and integrate third-party services into the platform. This experience gave me a strong foundation in designing scalable APIs, working with production data, and building features that must be reliable and maintainable in a large codebase.
In addition to my professional experience, I have built several personal projects using Python and FastAPI that involve API design and data integration. For example, in my Gather Google Calendar agent project, I developed a backend service that integrates with external APIs to help users make scheduling decisions. The system uses multiple third-party services as tools for the agent and exposes functionality through a FastAPI server that communicates with a Google Apps Script frontend. This project required designing clear API boundaries, handling external data sources, and building a service that could reliably support automated workflows.
I have also worked on infrastructure-oriented projects such as an LLM inference service, where I implemented batching and observability features using tools like Prometheus and Grafana. This experience gave me exposure to collecting and analyzing system-level metrics, which is closely related to the goal of improving the Dashboard’s system-wide metrics.
Together, these experiences have given me practical experience with backend development, API design, and working with data systems, which I believe will translate well to implementing new metrics endpoints and data download capabilities for the Programs & Events Dashboard.
Describe any open source projects you have contributed to as a user and contributor (include links).
I have gotten involved in open source software in the last couple months and have been contributing to projects I find interesting or useful. Here's a summary of my main contributions. All contributions are also visible my GitHub.
Balancer project:
I am currently a volunteer contributor for a civic software project that recently reached its MVP. The project's tech stack consists of React, Postgres, and Django. My work has ranged from small bugfixes, refactors, and documentation updates to building features, scoping & iterating on project design with the rest of the team, and conducting code reviews for other developers.
//mem0://
While learning about memory management for AI agents for a personal project, I found mem0--a personalized, intelligent, open-source memory layer for AI agents. I've made a couple contributions to the project ranging from small bugfixes to features. I hope to continue contributing to this project as it grows, since the challenge of building a long-term memory layer is one of the most interesting aspects of agentic AI software to me.