We just open-sourced one of the more interesting tools we have developed for our design partners - deterministic rule-based retrieval in RAG. Developers frequently tell us that they know exactly where to find the answer to a question within their raw data, but for some reason, their RAG solution is not pulling in the right chunks. To help with this, we are open sourcing a rule-based retrieval solution whereby developers can define rules and map them to a set of chunks they care about, giving them more control in their retrieval workflows. Check it out on Github (https://github.com/whyhow-ai/rule-based-retrieval) and please help star 😛

I helped build this, so am happy to answer any questions or take feedback about it too

Very well done y'all -- this is super cool

Rule-based RAG can efficiently handle growing datasets by filtering relevant context, reducing computational overhead. Targeted retrieval allocates resources effectively, delivering high-quality responses across increasing use cases for "gpu poor" enterprises who are looking for the best bang of the buck Also as gen ai use cases increases we are likely going to see higher percentage of cheap and synthetic text As data explodes, highly sophisticated rule-based RAG would be a norm as opposed to vanialla ones