gruporaia/Chat-Diario-Oficial

Chatbot baseado em IA que usa RAG para buscar e gerar respostas sobre documentos como o Diário Oficial.

18
/ 100
Experimental

Implements a two-stage pipeline: data ingestion via Apache Airflow orchestrates daily PDF-to-text conversion and semantic chunking (2000 chars) from the Querido Diário API, with embeddings stored in FAISS; the QA stage enriches user queries to boost domain terms (article numbers, agency names), retrieves similar chunks, and generates answers via Cohere's command-r-plus model with automatic validation and context expansion. Integrates MinIO for object storage, LangChain for LLM orchestration, and uses Cohere's multilingual embed-v2.0 for vector representations.

No commits in the last 6 months.

No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 4 / 25
Maturity 1 / 25
Community 13 / 25

How are scores calculated?

Stars

5

Forks

2

Language

Python

License

Last pushed

Feb 25, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/gruporaia/Chat-Diario-Oficial"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.