Gen AI Document Summarization and QA

I need a Generative AI model to summarize bank-related documents like policies, notices, and guidelines. The model should be implemented on Google Colab Free using "ibm-granite/granite-3.3-2b-instruct" using "HuggingFace" and "Langchain" Framework.

Requirements:
- Utilize FAISS, RAG for QA
- Implement map-reduce for summarization whenever user input query has keywords like ["Summary","Summarize","Brief","Overview"]
- Create a Streamlit UI with:
- File Upload
- Interactive Text Box

Functionality:
- User input to determine action
- Keywords for summarization: "summary", "summarize", "overview", "brief"
- Output format: Text or Markdown, whichever is good for UX

Ideal Skills:
- Experience with Google Colab free
- Familiarity with Generative AI, IBM models, HuggingFace, Langchain
- Proficient in FAISS, RAG, and Streamlit
- Strong background in map-reduce techniques