rag-demystified VS waggle-dance

This was exactly my experience trying to write a RAG pipeline that answers complex questions by generating sub-questions [1]. The hidden prompts in the complex libraries are quite difficult to get to and then start tweaking, whereas writing my own pipeline boiled down the code to 4 LLM calls.

[1] https://github.com/pchunduri6/rag-demystified

This seems very similar to LangSmith’s trace monitoring, which I have been leaning on heavily for observability. You also mention LlamaIndex— how do you see your project fitting into the ecosystem?

This is a great README, but I don’t think I would able to use this because it is serial.

In my experimental agent system, waggledance.ai, I have been working on a pre-agent step of [picking and synthesizing the right context and tools](https://github.com/agi-merge/waggle-dance/blob/main/packages...) for a given subtask of a larger goal, and it seems to be boosting results. It looks like now I have to try sub-question answering in the mix as well.

This is amazing. I will try to have it automate my system of agents web app (turtles all the way down) (shameless plug: https://github.com/agi-merge/waggle-dance)

- source code: https://github.com/agi-merge/waggle-dance

Any advice for trying to implement this in my project over at https://github.com/agi-merge/waggle-dance

Currently I am creating different agent types for planned subtasks using langchain, so perhaps implementing a custom AgentExecutor? Or would I need to lift it up higher in the logic stack? I am not sure that I understand how the graph search and thought-action-reflection selection process is deciding when and how to reflect if a branch fails, and how it backpropogates the failure to other nodes?

I have been working on a system of agents over at https://github.com/agi-merge/waggle-dance - I already split problems up into subtasks for agents to work on independently. I give agents access to vector databases, using a simple global key for now, but soon a context/parent/child key. Access to the vector DBs is proxied via tools (agents have to “call” saveMemory or retrieveMemory). I also check for looping/repetition FREQUENTLY using in-memory vector databases of the langchain agent callback events.

My opinion on this: eh, who cares? AutoGPT and similar are non-standard use cases for Vector DBs right now, and Vector DBs are useful for RAG.