.@karpathy's autoresearch gave us an idea. Instead of using the pattern to optimize ML training, we pointed it at a production RAG knowledge base to autonomously fix retrieval quality. 8 runs, $88 in compute, pass rate from 22% to 89%. The pattern works well beyond ML.

Swenor Consulting
@SwenorHQ
03-25
We took Karpathy's autoresearch pattern and pointed it at a RAG knowledge base instead of an ML model. 8 runs, $88 in compute, pass rate from 22% to 89%. The pattern is not about ML training. It is about disciplined autonomous optimization.

Sector:
From Twitter
Disclaimer: The content above is only the author's opinion which does not represent any position of Followin, and is not intended as, and shall not be understood or construed as, investment advice from Followin.
Like
Add to Favorites
Comments
Share
Relevant content




