How satishk01/llmsample works
Project type and purpose
• At the top level this repo (“LLMsample”) is an open-source collection of small demo apps and notebooks illustrating LLM integration patterns (AWS Bedrock & Kendra RAG, Huggingface image-to-HTML, SQLite-backed chat agents, local Llama2 workflows, PDF ingestion, etc.).
• Each subfolder (e.g. Claude3KendraRAGBot, bedrockagents, imagetoHtmlCode-Huggingface, llama2kendrarag, etc.) is a self-contained demo that shows how to wire up data sources (PDFs, images, SQLite, JSON APIs) to an LLM endpoint.
File structure
• Top-level directories = individual sample modules.
• Common file types:
• Python scripts (app.py, genIndex.py, kendra_bedrock_query.py)
• Jupyter notebooks (.ipynb)
• Lambda code (Bedrock-agents DB lambda in JS)
• Config files (.env, requirements.txt, .readme.txt)
• Dependencies: AWS SDK calls (Bedrock, Kendra), SQLite, Huggingface transformers, local Llama2 container.
Readme insights
• README.md simply states “Sample code for LLM,” so we rely on folder names and file names to infer data flows and external integrations.