Inspiration
I want to gain more experience with Azure ML service and vector offerings to expand my skills and knowledge in building AI-powered applications.
What it does
The project aims to create a custom AI copilot that can answer user queries by leveraging the capabilities of vCore-based Azure Cosmos DB for MongoDB and the Azure OpenAI API.
How we built it
I built my AI application using the Azure OpenAI API and vCore-based Azure Cosmos DB for MongoDB. I explored features leveraged by ChatGPT and studied capabilities, best practices, and real-world use cases such as vector databases and Retrieval-Augmented Generation (RAG). By completing the challenge, I developed a custom AI copilot that showcases the skills and knowledge acquired throughout the process.
Challenges we ran into
I encountered some challenges with the provided GitHub code samples during development. Some of the logic in the code was outdated, so I had to modify and update it as I progressed based on our learning and understanding of the current best practices.
Accomplishments that we're proud of
I'm proud to have successfully built our AI copilot, demonstrating our ability to leverage cutting-edge technologies like Azure Cosmos DB and Azure OpenAI API to create intelligent applications.
What we learned
This project taught us how to build an end-to-end copilot application using Retrieval-Augmented Generation (RAG) by utilizing Azure Cosmos DB. I have gained valuable insights into the process of integrating various Azure services and applying best practices to create a functional and efficient AI-powered application.
What's next for Phase 1 : Co-Pilot Submission
Moving forward, I would like to extend the capabilities of our AI copilot to create a research buddy with an advanced RAG approach that can assist users in their research endeavors by providing relevant information, suggestions, and insights based on their queries and context.
Built With
- azue
- cosmosdb
- python
Log in or sign up for Devpost to join the conversation.