Inspiration I wanted to utilize my learning on generative AI and prompt engineering to build a prototype. I decided to use Gen AI to build an interesting application out of the box, but I didn't get much time to think about it due to other work. However, two days ago, I sat down to think about what the quickest project would be to build using AI, and then I got the idea to use Vertex AI and a provided dataset to build something. So, I decided to create a Tip Generator prompt or something like that. Hence, I started and was finally able to complete it after a lot of errors and struggling to sort them out.
What it does This is a simple baseball information prompt based on a provided dataset. Whatever question I ask regarding baseball, it will give relevant answers in short.
How we built it I built it using Python, Flask, HTML, CSS, JavaScript, and Vertex AI (for providing the query from the sample dataset to train the AI based on the dataset).
Challenges we ran into I faced challenges such as how to run the app and deploy it on Google Cloud using Docker and containers. Because this was the quickest way to deploy with no prior knowledge. after almost continuous countless hours finally able to deploy the application. Since I am not using any json or node.js, so firebase hosting failed for me.
What I learned I learned how to use Vertex AI by providing some tuning to get the desired output based on the provided datasets and how to create an application like an AI prompt. Apart from this, I learned how to deploy an application on Google Cloud. I am still facing challenges as I am new to this, but I learned most things by referring to Google resources and debugging while working on it.
What's next for MLB TIP Generator I have learned how to deploy my AI web application to Google Cloud. Next, I will try to create some other interesting projects using similar concepts and different ideas, utilizing Firebase and Flutter for Android and iOS applications.
Log in or sign up for Devpost to join the conversation.