Thank you for participating in the RAG and Agents Bootcamp! To receive your certificate of completion, please submit two projects as outlined below. This assignment will demonstrate your understanding and application of the concepts learned during the bootcamp. We look forward to reviewing your innovative projects!
Assignment Question
Question 1: RAG Pipeline using BeyondLLM
- Create a Retrieval-Augmented Generation (RAG) pipeline using BeyondLLM.
- This project should demonstrate your ability to integrate various components to retrieve relevant information and generate coherent responses.
- Ensure that your submission includes comprehensive details and documentation that showcase your approach and methodology.
- Additionally, the notebook should contain the implementation output, and a deployed Streamlit or Gradio or deployed link should be provided below the notebook.
Question 2: AI Agent using OpenAGI
- This project involves developing an AI agent using OpenAGI. The focus is on creating and deploying the agent with a strong emphasis on its interaction capabilities.
- You have the flexibility to choose between different language models and configurations based on your access and requirements:
- If you have access keys to OpenAI, Claude or Azure LLM: You can utilize workers in your project to enhance the capabilities and scalability of your AI agent.
- If you do not have access to these keys: Gemini can be used for single worker while Ollama or Groq LLM can be used for single agent execution.
- Make sure the Agent output is visible in the notebook.
The RAG Challenge and Agents Challenge are still live! The top 3 notebooks from each question will receive swag, delivered to winners within India.
Guidelines
Register for this challenge using the same email ID you used for bootcamp registration, as it is crucial for receiving your certificate.
- Combine the code from Question 1 and Question 2 into a single notebook, ensuring to include clear sub-headings within the notebook.
- Make sure to add the deployed Streamlit link for the RAG application, and ensure the output for the Agentic workflow is visible.
- Once your code is ready, click on "Make Submission."
- Title your submission as
_assessment. - In the summary section, briefly describe the approach you took.
- Don't forget to tag it as deep-learning.