No time waste

Telegram Bot
Prompt engineering
Prompt engineering

Inspiration

Throughout the world, every single year, 2 billion + birthday cards are sent. Well, that may be quite less considering the total number of people living on earth but we need to consider the fact that birthday cards aren’t popular everywhere in the world. We take a lot of time to generate happy birthday wishes and gifts. So I took the chance to create this telegram bot to help us decrease this time. There are many birthday cards which when opened, play a song or music. These cards use far more computing power compared to what was actually used for sending a man to the moon! So here I want to solve this problem by integrating the mere concept of birthdays with bots.

What it does

It generates wishes and potential gift items based on a pre-trained cohere AI model. So when you get started with the bot it will greet you and allow you to click to help and see the potential commands. After that, when you click the generate it will generate the wish and gift for you.

How we built it

I used the Cohere AI model to train it using NLP to generate those wishes and gifts. Using the Cohere platform to build natural language understanding and generation into your product with a few lines of code. Cohere’s large language models can solve a broad spectrum of natural language use cases, including classification, semantic search, paraphrasing, summarization, and content generation. Through finetuning, users can create massive models customized to their use case and trained on their data.

Challenges we ran into

I ran into some challenges as it was my first time developing a bot on python. Specifically, I had a hard time fetching the data from the API and displaying it to the telegram bot. I also couldn't find teammates as I had the worst schedule this weekend but managed to not pass my very first MLH birthday.

Accomplishments that we're proud of

I'm so happy that I made this and I will continue to work on it as it helps a lot. I also faced some personal problems and had to work hard until the last minute and I'm happy that I'm done on time.

What we learned

I learned a lot about prompt engineering, cohere ai models, and telegram bot. I also learned some principles and techniques for writing prompts (inputs for our models) that will help you get the best generations for your task. Choosing the right temperature can also have a big influence on generation quality

What's next for No time waste

I will add more functionality to it and hopefully make a discord bot as well. One of the functionalities is to directly send the wish generated by Cohere to send it to the person without copying and pasting. Apart from that, I want to add some prices and images associated with the Gift so that it can let the user have a better understanding.