In this recording, i show a more complex implementation of llamaindex workflow
https://youtu.be/LloUNBD9fsI

I liked the way llamaindex is dedupliacting the nodes in the KG. In this recording, I explain in details how it is being done
https://youtu.be/vMz0icWZd5A

Will we be able to run multiple copies of the AgentService in a container, each having a shared state

It also looks like the orchestrator is LLM based, so there will always be a chance of non-determinism. Is that understanding correct as well

I have one question regarding llama-agent. I looked into the code. It looks like the orchestration and the agent services are tightly coupled. So, if I have to use the Agent Service in aother orchestration, I will have to duplicate the Agentservice and the processing of the agent. is that understanding correct?