We have a goal: Bridge innovation toward development of work talent and state of the art AI apps.
What it does
Git-Soft Engine is a platform designed to connect developers with open-source projects that align with their passions and expertise.
- Personalized Recommendations: Receive project recommendations based on your skills and preferences.
- Advanced Matching: Our engine considers factors like programming languages, project size, and popularity to provide tailored suggestions.
- Project Discovery: Explore open-source projects, view detailed descriptions, and access project repositories directly.
- User-friendly Interface: Enjoy an intuitive and user-friendly design for easy navigation and collaboration management.
How we built it
We built this project into Frontend, Backend and Data.
Frontend Streamlit was implemented as the framework to turn our Python code into visual and interactive elements.
As external (not built-in) components, we used the following ones:
Backend We communicate with Softtek SDK to get recommendations based on costumized data and the integration with OpenAI ChatGPT.
Additionally, we performed a data preparation step using over 2.8 million open GitHub repositories. This is explained ahead.
Data We connected to the GoogleCloud BigQuery server to access the public data from big query datasets. We get the GitHub repositories database that contains data from over 2.8 million repositories from GitHub.
Finally, we extract a sample of the big query dataset and preprocessed it.
Challenges we ran into
- Frontend: Design concept for UI/UX.
- Backend: Pinecone vector Database.
- Data: Google Cloud, Big Query and Python configuration.
Accomplishments that we're proud of
- Suitable MVP for Demo.
- Big Data pre-processing.
- Architecture integration.
What we learned
- Frontend: How to use Streamlit.
- Backend: How to communicate with Softtek SDK.
- Data: How to manage Big Data using Google Big Query.