Our project gemini chatbot was inspired by the growing enthusiasm for conversational interfaces powered by artificial intelligence and our curiosity about what can be achieved with Google’s Gemini platform. So, we set out out goal to create a chatbot tool that would use Flutter and the Gemini API to simulate human-like conversations.
During the development process, we learnt about natural language processing, machine learning algorithms and how difficult it is to build conversational AI systems. We aimed at an intuitive chat interface that was also dynamic, this would require us to work with Flutter’s flexibility as well as some of GCP’s features while developing on VS Code.
For this project, we had to connect the user front-end in Flutter with google's back-end through an API called gemini which acted as a translator between them. We trained models provided by gemini so they could understand different responses better than they would have done initially then personalized those models according to our project needs, additionally besides other things like context awareness or sentiment analysis which added richness into making it more interactive conversationally speaking.
Though it was thrilling working with advanced technologies, there were many challenges faced throughout its execution because certain aspects proved tricky. Some of them were technicalities such as integrating the Gemini API into flutter and deploying on GCP which required extensive troubleshooting through experimentation.
Built With
- flutter
- google-cloud
- vs-code
Log in or sign up for Devpost to join the conversation.