Learning how to answer support questions based on recent entries. Demonstrating the difference between RAG Disabled and RAG Enabled modes. Generating support tickets based on the style of highly rated questions in the database.
Importing recent Stack Overflow data for certain tags into a KG. Embedding questions and answers and storing them in vector index. Choosing tags, running imports, and seeing progress and some stats of data in the database.
Loading a local PDF into text chunks and embedding it into Neo4j. Asking questions about the PDF contents and having the LLM answer them using vector similarity search.
Building the application separately from the back-end code using modern best practices. Instant auto-reload on changes using the Docker watch sync config.
