alejandrorodriguezdelarosa-ux/asistia-rag-agente-conversacional

AsistIA is a conversational RAG agent for Generative AI: a demo/advisor built with Google Gemini, ChromaDB, LangGraph/LangChain and persistent memory. Intended for students and developers (final project for a Master module), it answers Spanish queries grounded on a local vector knowledge base (three PDF docs) and keeps per-conversation memory. Key features: chunking via RecursiveCharacterTextSplitter, embeddings with Gemini, ChromaDB persistent collection (./chroma_db, top-3 cosine retrieval), MemorySaver-based thread memory, system prompt enforcing use of retrieved context and Spanish output, explicit fallback when context is missing, low temperature (0.3) for consistent technical answers. Repo includes a Jupyter notebook, optional Streamlit app, requirements and .env template.

AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.