Skip to content
A

AzaryaaAI Bot

The AI-powered crypto chatbot enables 24/7 global support, delights users, and creates more time for human team members to focus on core project priorities.

Created on 25th July 2023

•

A

AzaryaaAI Bot

The AI-powered crypto chatbot enables 24/7 global support, delights users, and creates more time for human team members to focus on core project priorities.

The problem AzaryaaAI Bot solves

Cryptocurrency projects need 24/7 global community support but human moderators cannot cover all timezones, know every answer, or scale conversations as adoption grows. This leads to unanswered questions, incorrect information, frustrated users, and moderator burnout which further leads to poor user experiences.

The onboarding process in Web3 is broken. When someone new to the Web3 world visits a project website, there is a high chance that they will not understand anything. The terminology, the jargon, and the overall experience can be very daunting for newcomers. This bot helps to make onboarding easier by providing clear and concise instructions, as well as by answering the queries.

It automatically absorbs knowledge from the project's whitepaper, website, blog posts, and other materials, as well as through moderator inputs.

This bot solves the problem of inconsistent, limited-time community support. This results in happy users, reduced workload for humans, and accelerated project growth.

Challenges we ran into

We spent a lot of time in learning about the generative AI.There are not a lot of strctured courses on the topic so we found it a bit diffulcult to build the architure of the project.
Once we had a basic understanding of generative AI, we began to work on the architecture of our project. We knew that we wanted to use a large language model (LLM) to generate text, but we were not sure how to integrate the LLM with our existing stack. We eventually decided to use the Transformers library, which provides a simple and efficient way to use LLMs.

After the LLM stack was finalized, we had to find a way to send the documents of the project to the LLM stack. This was a bit difficult, as our entire stack was in Python and the uploaded documents were on Firebase. We eventually solved this problem by using the firebase-python library to interact with Firebase.

We are still in the early stages of development, but we are excited about the potential of our project. We believe that generative AI has the potential to revolutionize the way we create and consume content. We are committed to continuing to develop our project and to making it available to the public.

Tracks Applied (1)

Comunity

We are building this tool to make the community tool more delightfulThis project centers around building an intelligent ...Read More

Discussion

Builders also viewed

See more projects on Devfolio