Deploy RAG System with Featherless AI

Describe the steps you would take to deploy your LlamaIndex-based RAG system as a scalable, production-ready API using Featherless AI. Focus on how you would containerize your application, configure Featherless AI for inference serving, and expose the agent's chat endpoint. Include considerations for managing `GPT-4o` API keys and `Pinecone` credentials securely in a production environment.