This repository serves as an accompanying guide of hands-on examples to a AWS startup loft session delving into canonical generative AI use-cases.
-
RAG (Retrieval-Augmented Generation): This folder contains a detailed guide to RAG, one of the critically acclaimed methods for generative AI. It incorporates instructional code snippets, alongside modern RAG techniques (rewrite-retrieve, HyDE...)
-
Summarization: The summarization part provides experiential learning by walking you through a series of pilot projects which tackle real world problems using AWS and Chain of Density method.
-
Text to SQL: This section handles language understanding with LLM, specifically looking at text to SQL translations. This use-case explores the potential of conversational AI models, especially using ReAct framework.
The use-cases in this repository are scripted as Jupyter notebooks. You can either clone the repo to your local machine and follow along, or run the notebooks directly on SageMaker Studio.
Your contributions are more than welcome - let's grow together! Feel free to clone, modify, raise issues, or suggest enhancements.
I am deeply grateful to DoiT International for creating an environment conducive to learning and growth. Also, a huge shoutout to the open-source community and AWS for making exceptional knowledge resources accessible to everyone.
For collaboration, guidance, or questions, kindly refer to the 'Contact' section of my portfolio or drop a message via LinkedIn