I've built a RAG app. now what?

1 points by curiousbear 1y ago ↗ HN

i've built an advanced RAG app which works on top of your Google Drive. can help you query your documents in near real-time with a hybrid search functionality. but i see that many companies are working on similar systems. what could be some good use-cases to differentiate our solution? where we can potentially build some moat in long term

2 comments

[ 4.1 ms ] story [ 17.1 ms ] thread

verdverm 1y ago ↗

AI is not a product, nor is RAG, they are at best algorithms. There is no moat at this low level, the field is rapidly evolving.

To differentiate, build a product for an industry that includes AI/RAG features. This requires expertise and experience in said industry to build something useful or competitive.

papadany 1y ago ↗

The vector DBs and companies like OpenAI will eat up such small solutions very fast. They all have a great advantage already in the RAG space. So, better to think about creating something else other than building yet another RAG solution. But I feel like there is still a lot of gap in evaluation part, so watch out for that space.