Custom RAG Pipeline for Biology Textbook

I'm in need of a skilled individual who can design a custom RAG (Retrieval Augmented Generation ) pipeline to assist in answering short answer questions from the openstax Concepts of Biology textbook.

- The RAG pipeline should be able to answer a variety of short answer questions from the textbook.
- Detailed explanations for the answers are required, as they will be used to check the understanding of the material.
- The ideal output for the answers provided by the RAG pipeline should be of moderate accuracy, acknowledging that some errors are acceptable.

It's essential that you have experience in designing and implementing RAG pipelines or similar text analysis systems.

Important Pointers:
1. Download the pdf from the link above
2. To make indexing faster, you can pick any 2 chapters from the pdf and treat it as a
source.
3. Use any in-memory vector database if required.
4. Use any open source HuggingFace model as the LLM Model
Output expectations-
1. Colab notebook to run the backend logic and evaluations:
Please add text blocks in your Colab to add scenarios/assumptions etc to make it readable.
Additionally if one can produce alternative way to do the RAG without using any library like Langchain,
LLamaIndex or Haystack