SocialAI

• 

  SocialAI - Your AI Bestie for Fantastic Friendships

• 

  Activities Suggestion Feature

• 

  Intelligent Message Writer Feature

• 

  Workflow in OutSystems

Inspiration

The inspiration behind SocialAI is to improve the relationship with friends, especially for those who are busy with work and have forgotten to mingle with their friends. The idea is to use the app to remind the user when his friend is having a birthday and suggest activities to do with their friends based on their activity feed on Instagram.

What it does

The app allows users to connect with their friends on social media, such as Instagram and Facebook. It can remind users of upcoming events such as birthdays and suggest activities to do with their friends based on their hobbies and lifestyle. It also has a feature to help users write the perfect message for their friend based on their past messages.

How we built it

We built the application using OutSystems, a complete full-stack application development platform. We used the platform to quickly create mobile and web applications, chatbots, and reactive web apps for any device. We built the data models, workflows, logic and pixel-perfect user interfaces and interactions, and even added our custom code. We also used a low code platform to develop a module that talks to the Gemini API instead of using the Python library.

Challenges we ran into

One of the biggest challenges was ensuring the app was able to retrieve medias from social media platforms. We tried to use the Instagram graph API, but the computation was heavy and we needed to host a backend to perform this specifically. We also ran into challenges when making Gemini understand the image and reply in the format we wanted, so that the LLM output could be used by the application.

Accomplishments that we're proud of

We are proud of the fact that we were able to build a functioning app that allows users to connect with their friends on social media and be reminded of upcoming events. We are also proud of the message writing feature, as it allows users to write the perfect message for their friends based on their past messages. Additionally, we are pleased that we were able to develop a module in the low code platform using dotnet to talk to the Gemini API instead of using the Python library.

What we learned

We learned the importance of hosting a backend to perform heavy computations. We also learned the value of the low code platform and the ability to easily develop modules using dotnet and link them to the Gemini API. We also learned the value of understanding the language nuances And training the LLM to to generate the perfect message.

What's next for SocialAI

Looking to the future, we plan to expand the app to include other social media platforms and improve the message writing feature to make it more user-friendly. We also plan to add more features such as the ability to schedule events and plan activities with friends. We also plan to deploy the backend for heavy computation and improve the performance.

Built With

• gemini
• mysql
• outsystems