BeaGL is an app that was created as research project under UC Davis professor Sam King. Our solution seeks to utilize the advents of machine learning and modern Continuous Glucose Monitors (CGMS) to create app an app that addresses tighter glucose management.
During the process, I led the overall design and front-end development of BeaGL.
SKILLS
Product Design, Front-End Development, Brand Design
TIMELINE
2 months
TEAM
Sriram Magesh, Jun Min Kim, Brendan Leung, Tim Stewart
Living with diabetes means constantly walking a tightrope between stability and uncertainty. Every decision from what to eat to how much to run has to be accounted for to prevent hypoglycemia and hyperglycemia. To address this, four key areas were identified to help create the MVP for BeaGL.
Reducing Cognitive Load
Diabetics are constantly required to do mental calculations on how the food they eat affects their overall health. What are some ways to allieviate to mental load?
Preventing Sleep Loss
Many with Type 1 Diabetes receive notifications from their CGM's during the middle of the night to alert highs or lows. Could this be avoided?
Improving Glucose Management
Tighter managment of a person's blood sugar levels is coorelated with a longer and healthier life. How can this app influence a person's control of their glucose levels?
Ensuring Security and Safety
To entrust a person's lifestyle on an app requires incredible attention to detail and technical development. How can the design of this app allow for a safe experience?
Our first step as a team was to conduct literature analysis on research papers and products that sought address the topic of glucose management. We were able to find similar products that would do certain things well, but couldn't check upon all the boxes that we wanted to be filled.
When thinking of potential solutions to address our product goals, the user flow that I had intially created contained too many layers and too many screens that the user would have to go through. Through discussion with our team, I was able greatly refine and simplify the experience to allow for max 3 interactions for user to access the home page.
This user flow went through many iterations. We had a realized as a time that the only form of verfication that was needed was just the person's phone number. Nothing else, no username, no password.
Our team together drew up an action plan to tackle how this app would be coded, and the manner in which information would be stored. Our goal with this was to store as little information as possible.
Initial iterations of BeaGL were a bit overwhelming in terms of number of screens needed to reach the home page. We worked on as a team cutting out pages that were unecessary.
While AB Testing, we found that a darker colored heading was a more visually appealing choice in comparison to the previously lighter version. Additionally, attaching confirmation buttons closer to input buttons was found to be more beneficial when typing an input.
Once the intial design for the MVP was complete, my role transitioned towards leading this app's frontend development. The language that our team used for this app was React Native.
Switching to React Native
This was my first time using React Native. I had prior experience with React, so I had used the foundation of the latter language to quickly learn Native.
Practicing good code writing
Developing for this app was good practice towards writing good code. It was my goal to use proper naming conventions, and commenting context of any code written.
Testing the device locally
Initially, I found it difficult to test the device locally on my phone. I kept running into many error messages due to packages that I had not installed.
Merging branches on Github
Most of my prior projects were solo, so this had been my first real expericence in merging branches on github.
The verification process for BeaGL involves the user inputting their phone number, and responding to a push notification.
Onboarding for BeaGL allows users to customize their preferred alerting ranges, and their secondary verification source.
The BeaGL homepage allows for users to view their Glucose readings and its predictive trend. The person also has the ability to view past readings and see any signs of improvement and progress.
The team and I are working towards updating and adding more features onto BeaGL. The app is currently being used by 6 UC Davis students, and has a plan to expand to the wider public. It's been such a joy to work with this team, and build on this amazing product, and I'm excited to see where it goes!
BeaGL team @ Zion National Park
Feel free to contact me for any future projects.
