Aritras-Colab
Beginner-focused learning repository and Jupyter notebooks for GenAI & RAG pipelines.
Technical Case Study
Designed an educational codelab repository. Notebook 1 covers Google Gemini API configurations (temperature, chat sessions, embeddings, safety, and streaming). Notebook 2 details a complete end-to-end RAG pipeline: document loading, chunking (RecursiveCharacterTextSplitter), embedding (gemini-embedding-001), vector database storage (ChromaDB), and inference using Google Gemini-2.5-Flash.
# Run traditional RAG pipeline test
> Loading document chunks from dataset: Aethelgard.txt
> Vector embedding: gemini-embedding-001 (1536 dim)
> Storing chunks in ChromaDB vector repository...
> User query: "What are the defense coordinates of Aethelgard?"
> LLM Response: "According to the local archives, Aethelgard's defenses..."