Inspiration
Customers coming to restaurant always gets the same dish, they don't get variations. For restaurants, experimenting with food is the cheapest thing to do among other options.
Providing an option to customers for preparing the recipe will add personalization. It will also make customers feel that the chef is going to cook the dish proposed by them.
Experimenting with such dish could be risky and possibly operational nightmare for restaurants. The risk could be reduced by generating lot of preorders. Before committing to the dish, restaurants can safeguard themselves by having minimum order cap.
The inbuilt virality for each dish - customers will promote the dish in there network and social media so that they can ask Chef to cook it. The more people share it, more possibility of generating order.
Customers also get the chance to play some small fun game while they are waiting in the restaurant.
What it does
Restaurant admin will authorize the app, post authorization, admin of restaurant will copy the map link / print the QR code. This link will be shared on social media by restaurants, they can keep printed QR codes on each of the tables and give name as Bhaari Rides.
Customer coming to restaurant or anyone clicking the link will get a chance to register and ride the bike. Now each player will collect multiple stars - each star is mapped to random produce item e.g. carrot, cabbage, cucumber etc. Once player is able to complete the map successfully, they will get new randomly generated recipe using Generative AI.
Customer can place an order for this recipe or customer can share a link to this recipe on their social media. One dedicated link is created for newly generated recipe. Anyone who wants to have this dish, they can place an order, they can subscribe to this dish.
How we built it
First we created a framework to create custom maps - any way we want. Then we created couple of maps which will be long enough to keep the players engaged.
When customer registers for the game, we are creating corresponding customer on square.
Then we integrated all the raw materials e.g. carrot, cucumber, cabbage etc. and allowed user to randomly earn these items while they play by riding on the bike. At the end of map, we are taking all the items collected and generating the new recipe using OpenAI chatgpt apis.
We are also trying to generate the dish images using Bing image creator.
Once the dish is generated, customer will have an option to place an order, share it on social media. The new catalog item is created with random price amount. This item is then used to place an order for given customer for this new catalog item. If user has selected the subscription then we are generating subscription. Invoice is generated at the end.
Challenges we ran into
Making multiple components run in sync is challenge. Faced lot of challenges to make this functional end to end the way I have imagined. Still few places remaining where finishing touches are required.
Integrating Square APIs through Unity based game, putting all things together conceptually in the form of game was challenging.
Making SaaS kind of application on top of game by consuming Square api was challenging. Our first design to use the tokens were working well then we realized multiple restaurants will also use this app so we had to change the design, permanently store the tokens for future use.
Without having a verified account, we could not test the invoices and payment part.
Accomplishments that we're proud of
We are able to build a very complex system which helps restaurants to get more business using latest technology and helping customers to get new way to place an order and get surprise dish every time they play the game.
This game will help great online experience and converted equally awesome experience offline, increasing business at the same time.
What we learned
Lot of new technologies - Unity, Firebase, Square app marketplace and its API endpoints, Generative AI API endpoints, AWS hosting etc.
Apart from technology, we learnt a lot of things related to restaurant industry. Their pain areas and how we can solve this apart from this project.
What's next for Surprise Me
- Payment integration
- Seamless integration and experience with newly generated recipe
- Create dedicated maps for each restaurant - custom maps
- Integrate restaurants actual menu and offer customers to place an order from existing menu
- Group bookings
Built With
- amazon-web-services
- chatgpt
- dall-e
- firebase
- html5
- node.js
- openai
- unity
Log in or sign up for Devpost to join the conversation.