Inspiration

We wanted to build a one-stop shop for all things remote-interactive. This app gives everything the convenience of Google Docs with its synchronized editing. No longer will you need separate pair programming apps for your IDE, pair editing apps for your Office Suite, or remote play apps for your couch coop games!

What it does

It combines screensharing with input injection. Hosts can share their display to clients who can then click on the display and type freely without blocking the host's input streams.

How we built it

Luckily, Bryan has a competency in networking. For the rest of us, our fortes lie outside that field. This meant we harnessed ChatGPT for project planning and scope refinement, Claude for assisted code generation, and GitHub Copilot for extra code triage.

Challenges we ran into

Naturally, when using models to generate code, they can tend to produce unnecessary bloat code. We had to take several optimization passes throughout the project to increase performance.

Also, initially, we struggled with fleshing out the project's feature suite, but all of that became clearer as we began the work.

Accomplishments that we're proud of

Creating an open-source streaming application with input injection is something we're very proud of.

What we learned

mss, pyautogui, and a lot of the other libraries we used on this project were entirely new to us. Getting to experiment with input injection and screensharing was a very enlightening experience for the team. It has definitely given us a new, exciting lens through which we can plan future projects.

What's next for Pair Screen Sharing App

Decrease latency (increase client-side FPS), allow for remote network connections (currently, only machines on the same Wifi network can stream), add microphone audio streams (cut to reduce load on the application), and add a more robust profile system where users can have: bios, profile pictures, contact info, etc.

Special acknowledgements

As first-time hackathon participants, Bryan Bong and Christian Cardenas brought incredible passion, productivity, and dedication to our team. Their work was integral to Coniuncti’s development, and it’s clear the project would not have been nearly as rich or successful without their involvement.

Built With

  • claude
  • mss
  • opencv-python
  • pillow
  • pyautogui
  • pyngrok
  • pyside6
  • python
  • tkinker
Share this project:

Updates