Hi everyone. I am sure this is a very basic question but I haven't really found a good resource for solving my problem.

I have a bunch of structured data that I am currently able to query and perform well using GPTPandasIndex. However, what I would like to do is to build a text-based interface on top of this whose results in turn can be fed into some other part of the pipeline. Think about the following situation: Suppose my structured data is a massive inventory of objects. The user inputs a query such as Select 100 objects with the property that size of object is greater than 100mm . Once I have the output of this query, I want to run a python program on this output to perform some other operations. This could be something like. adding the output to a queue, and then using some other analysis on it.

So while I get the first part: getting the output from the query (I am currently using eval in python so open to better ideas), I want to be able to connect the query to other external python programs.

Hello everyone, this is not a technical question per se. But I am trying to build an internal tool for the startup that I work for. With the proliferation of LLM tools like Pinecone, Llama-index etc, its hard to see the forest for the trees. I guess my question is: How do the various tools fit with each other? For instance, how does Llama-index, a vector store like Pinecone, a tool builder like LangChain, and the LLM itself fir together.

I can move the question to another appropriate channel if the question is beyond the scope of discussion here. Thanks!!

So where does the index layer come in ? Maybe it’s a dumb question: but I just don’t … understand what an index is doing here. If the LLM can transform the query into a piece of code, then how does indexing help? I can then provide information about the data frame (like column names) in the prompt itself right ?