Skip to content

Ask HN: How to manage docs for LLM RAG app?

6 pointsJianghong944 comments
On HN

I'm building a LLM RAG QA bot for my company, a financial institution. Right now I know the 'basic' building blocks, e.g. prompt engineering, RAG, vector db, eval, etc. Funny enough the first challenge I encounter is to curate and manage all types of docs, e.g.: * email chains * teams recording transcripts * confluence pages * pdf manuscripts

These can be ever-evolving and may hook up with periodic delta updates, manual sync, add/remove, etc. And I'm trying to figure out if there's a way to manage these docs/texts properly. Basically, I think I would need a system to store these files, their metadata, etc, and provide a web UI for people to manage them. Then these blob of texts will go through frameworks like langchain/LlamaIndex and be cleaned/chunked into vector db, and different chunking strategies can be A/B tested while other people maintain this ever-growing docs system.

Any suggestions are welcomed. I've tried some all-in-one frameworks but so far my experience are lackluster. Also, my company due to compliance constraints cannot use cloud-based solutions, so it has to be either open-source local-deployed, or developed locally.

Comments

Did you try dify? I found it was a good beginning for me.

https://dify.ai/

Yes. There're a couple of problems I encountered during my trial and ended up choose to build my own: 1. the version I was using requires the configuration of a 'reranking model', which AFAIK has to be from cohere. Funny enough, I can later opt to not use this reranking model, but I cannot get any functionality working if I don't preconfigure the cohere API key for this model. The problem with this is that, I need to request my company's infosec to allow API call onto cohere, for a functionality that I didn't intend to use.

And I think this results in my observation of the startup llm app world vs the heavily-regulated finance industry world: that the startup llm world is taking a default opt-in, default saas mode; however the regulated clients cannot benefit from these practices due to arcane regulation requirements. In this case, dify requires the setup of an API key for a service that I don't intend to use, and I have to go through the hurdles just to get the hello world going.

My analysis is that, there are 2 ways to make the heavily-regulated clients happy: 1. major cloud provider integrating everything into the myriad of its service catalogs and offer a full suite of services with the sales and compliance completed as a whole. 2. startup go down a curated, opinionated, battery-included oss approach with the least of sass fuss.

2. for some reason when I test using the simplest QA template, it seems dify cannot retrieve the document of any kind that I uploaded. I don't know why it happened but debugging the simplest offering of an app that boasts ease of use is not my thing.

I didn't go further from there. I wish my experience is just an extremely rare case since dify definitely seems to be the most polished all-in-one provider out there. If my experience turned out to be more positive I'll definitely stick with it, now I just feel picking a all-in-one is not the right choice.

you own it dude? Damn crazy good UI

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.