Inspiration
Some of the team members meal prep for the next day and wanted a randomizer that would ensure that they would have healthy meals (sometimes).
What it does
The assistant asks the user for a calorie intake number and suggests breakfast, lunch and dinner based on that.
How we built it
Using Google's Dialogflow and javascript for the fulfillment part.
Challenges we ran into
The structure of intents are more targeted for scenarios where the user requests information and the assistant replies. In our scenario the tables are reversed. The assistant asks the user questions and the answer becomes the input to the next intent.
Accomplishments that we're proud of
That we were able to finish the MVP version we had in mind.
What we learned
How to train the assistant with variations of a phrase, set up events and create logic to branch the conversation.
What's next for Meal Plan with Google assistant
We would like to recognize meal types (i.e. vegan, vegetarian, healthy) and suggest meals that meet the criteria. We would like to link to an API that would provide us with meals and calorie counts so that the choices would be more varied. We would also add the ability to generate a new suggestion for a specific meal in case the user doesn't have the ingredients or he/she doesn't have the ingredients. Finally we would like to be able to send a text to the user with the final meal plan.
User Guide To Test the Meal Planner
- User: Create a Meal Plan / what should I eat tomorrow?
- Google Assistant: How many calories?
- User: My calorie intake is 1900
- Google Assistant:
- User: Thank you
- Google Assistant: Good bye!
Built With
- diagflow

Log in or sign up for Devpost to join the conversation.