Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
CuriouslyC
79 days ago
|
parent
|
context
|
favorite
| on:
The RAG Obituary: Killed by agents, buried by cont...
RAG isn't dead, RAG is just fiddly, you need to tune retrieval to the task. Also, grep is a form of RAG, it just doesn't use embeddings.
nbstme
79 days ago
|
next
[–]
Yes my point is that the entire RAG pipeline like ingest, chunk, embed, search with Elastic, rerank is in decline. Grep is far simpler. It’s trivial.
tw1984
79 days ago
|
prev
[–]
No, grep is not RAG. RAG is all about embeddings + vector search + LLM working under a fixed workflow.
Saying grep is also RAG is like saying ext4 + grep is a database.
rpcorb
77 days ago
|
parent
|
next
[–]
You don't decide what "RAG" means, a term that's been around for much, much less time than "database".
CuriouslyC
79 days ago
|
parent
|
prev
[–]
So you're saying grep isn't a form of information retrieval?
tw1984
79 days ago
|
root
|
parent
[–]
information retrieval is a much larger superset of RAG.
grep + agentic LLM is not RAG.
leobg
78 days ago
|
root
|
parent
[–]
Retrieval Augmented Generation. How you do the retrieving is irrelevant. You can do it manually and it’ll still be RAG. Also, most RAG pipelines combine multiple approaches - BM25, embeddings, etc..
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: