Technical Projects

Select project:

Or use the scroll-down markers on the right-hand side of your desktop screen.


PennApps XX | September 2019

  • I designed, developed and deployed our website using HTML, CSS, and Flask

  • I created the presentation and pitch, providing accurate statistics and data regarding immigration policies and their effects.

Reunited provides a platform for families to find their missing children.

The deportation and forced separation of immigrant families crossing into the United States has psychological effects on individuals and families and gives rise to a public health crisis that can affect entire communities, according to a new report published in the American Journal of Community Psychology. Children who lose a parent to sudden, forced deportation experience anxiety, anger, aggression, withdrawal, a heightened sense of fear, eating and sleeping disturbances, isolation, trauma, and depression. Children also experience housing instability, academic withdrawal, and family dissolution. Older children often need to take on jobs to help support the family. With families being separated by raids and children being inhumanely kept in detention centers or sent to far-flung locations, our team found it necessary to address this social injustice.

Reunited provides a platform for families to find their missing children as well as a way for concerned citizens to report a child's information to help find their family. The web app compares two photographs of children and determines whether those two children are the same person with an accuracy of over 95%. This creates a match between parent and temporary caretakers in order to reconnect the children with their families.

We used Amazon Web Service's Rekognition API to determine identity of the child regardless of factors such as lighting, time, angle, environment, or superficial changes like hair color. We also created a database using mySQL and hosted on AWS where we saved all the images and the information about the missing children and their parents. We put our algorithm and everything together via a Flask web app where we built the back-end using REST API’s in Python.


uw Energy Hacks | June 2019

  • I collected and cleaned the data

  • I edited the front-end code

My team and I predicted the difference in cost, revenue and environmental impact between renewable and nonrenewable energy sources in the hopes of incentivizing the industry to shift towards more sustainable energy sources.

Captain Planet

Dragonhacks | April 2019


🏆 MLH: Best Domain Registered with

  • I collected, analyzed, and visualized data used to create the environment of the VR game.

  • I created the presentation and pitch, providing accurate information regarding environmental sustainability.

Captain Planet is an educational virtual reality game to educate and empower the public to save our oceans.

Despite comprising nearly three-quarters of the Earth, the oceans have turned into large trash cans filled with human waste. Improper disposal of non-biodegradable substances like plastic, and the release of toxic compounds from agricultural runoff and industrial emissions severely influence the decline of water quality worldwide. This leads to the destruction of marine ecosystems that are essential for the survival of our planet and of humanity. Since education is the best way to empower the general public to take action to save our planet, our team created Captain Planet, a virtual reality educational game, to tackle this challenge in an interactive and accessible way.

Captain Planet simulates an interactive underwater VR domain to educate people of all ages about the effect that human activity has on the degradation of marine ecosystems. It takes the user in a multiplayer underwater ocean bed, whilst monitoring the health levels of the surroundings. It provides a closer look on water pollution, dead zones, and the basic steps that one can take to preserve oceans. It is a multi-level game that begins by restoring a stable ecosystem in a dead zone covered by algal bloom.

We used Unity with ARCore to host and develop our VR game. We used a custom network linking interface for multiplayer tracking and support. We used .cs C# scripts to develop and interact with UI elements. We also overlayed a world map with button controls to interact with the game. Further, the supplementary website (with more robust information, accurate data, and resources) was created using elements from HTML and CSS. The Firebase database and platform was used in multiple instances as well. The website loads the data from Firebase, as well as the VR elements that load data elements for status monitoring.


Techtogether Boston | March 2019

  • I designed and developed the front-end (HTML + CSS; Adobe XD wireframes) of the web app.

  • I also helped troubleshoot some of the back-end (node.js) , including the image recognition API and the primary on-boarding form.

MySympTracker’s homepage

MySympTracker’s homepage

MySympTracker uses image and emotion recognition to analyze, store, and track medical symptoms.

One of the biggest disparities in healthcare is the widespread lack of knowledge regarding many common health issues. My team wanted to create an application that would help address this and thereby bridging the gap between medical communication, making room for more universal access to healthcare.

Our application takes advantage of Microsoft Azure’s image and emotion recognition to track medical symptoms. Users can create their own accounts to save data taken from their sign-up form (including their name and contact information) and their follow-up questionnaires (including demographic information, previous medical history, and the contact information of their primary care physician) into a MongoDB database. Users are then able to utilize a service where they can upload a recently taken photo to the application, which will be displayed and analyzed using facial emotion recognition. Based on the most prominent emotion found on the user’s face, an additional questionnaire will be shown, detailing symptoms and health issues associated with the user’s displayed emotion. Those answers will be stored, and should the recordings pass a certain threshold and become unsafe or worrisome, an automated phone call will be made to the user’s primary care physician. Using the Twilio API, the call will be detailing the symptoms experienced over a span of time as well as the possible health issues they could be caused by, urging the physician to schedule an appointment with the user.

Our prototype could be improved with a mobile app for accessibility and convenience. A visual representation of the symptoms over time in an accessible way (eg, graphs) could also be beneficial. The team also hopes to see the image recognition used to help users better explain differences in symptoms like rashes or hives, so the physician is better equipped to treat the user remotely.



  • I drew the wireframes and created the UX design for the cross-platform mobile app.

  • I broke down the medical jargon into accessible language for both developers and users.

  • I also created the presentation and walkthrough demo.

Described as an American Crisis, the rate of maternal mortality has been steadily increasing for the past decade. The United States is the only industrialized country where the rates of maternal deaths are increasing. New Jersey’s maternal mortality rate (ranked 45th) is nearly twice the national average, according to the United Nations’ 2016 Women’s Health Report. Postpartum Hemorrhage and Hypertension are the leading causes of maternal deaths. Racial disparities persist. Women of color account for almost 60% of pregnancy-related deaths in the United States.

Currently, many women are unaware of the risks for maternal death within the 42 days post-childbirth. The New Jersey Department of Health believes that public education and awareness are critical steps in confronting this epidemic.

Our team has built an MVP of a cross-platform mobile app (created using Google’s Flutter and Material). Our app has three primary objectives: (1) help pregnant and postpartum women learn about their own risk factors, (2) track and self-assess their symptoms and catch post-birth warning signs, and (3) prompt action, inform them about care options, and reach out to emergency contacts when needed. The MVP has the current features: sign up and log in with email and password (database created using MongoDB); have the user answer some basic questions (form created using node.js) to understand their state (pregnancy, postpartum, etc.); an educational dashboard with recommendations and articles based on their answers to the initial questionnaire; a prompt for the daily check-in (push notification), which tracks the user’s mood, discomfort level, etc.; smart device integration (heart rate, sleep) with the Oura Ring (using pre-existing Oura API); and the ability to send a text (using Twilio) to family members in case of emergency or for accountability.

🏆 Grand Prize Winners

🏆 Grand Prize Winners

MamaCare helps pregnant and post-partum women understand and cope with their risk factors.

Our users are:

  • educated
  • tracking their symptoms
  • protected with an automated notification + call to their PCP.


Bytehacks | September 2018

🏆 Best Education Hack

  • I designed and developed the front-end (HTML + CSS; Adobe XD wireframes) of the website.

  • I contributed to the data analysis and visualization using R and Python.

Our team wanted to utilize open source data for greater transparency in the education sector. We aim to help inform parents, students, teachers, and other concerned community members about the reality of underperforming schools in the United States, starting with New York City. We want to show that education inequality is more than demographics and test scores -- we have a duty to provide every child with a safe, clean, quality public education. Our hack shows that we are failing in New York City.

Our hack is a website which provides graphic data and links to relevant datasets which are either hard to understand in their raw form or not easily found without sifting through pages and pages of data on We focus on the data that user interfaces like do not show upfront, such as violence in and around schools, dangerous lead levels in water, and poor physical infrastructure. We also provide functionality to find NYC petitions for education improvement, contact your local representatives, and donate to local schools. I focused on data science, finding datasets on and parsing and visualizing the data using Python, R, and Microsoft Excel.

KnowYourSchool visualizes open source data on the quality of NYC Public Schools.



As my first project outside of school, my team (ION SKY) created a mobile app prototype supplementing the NYC MTA that allowed users to view Planned and Unplanned Service Change Information and receive Real-time Arrivals & Schedules. Users could also report problems such as violence, uncollected garbage, or unexpected delays. With an account, they can collect points for their reports which could be claimed for a metrocard discount or towards an unlimited card.

This was created at least a month before the MTA first released their official MyMTA app, and nearly a year before it was actively used by the community.