Ask HN: When did we go from agentic loops to graphs?

#1

Post by grandimam » Sat, Aug 01, 2026, 5:23 PM UTC

The AI engineering conversation seems to be shifting daily: from prompts, loops, to now graphs.

I’m curious whether there is genuine progress happening in the field that has given me so much or whether folks are creating terminology because they want to feel closer to action or be able to say they predicted the next big thing.

Re: Ask HN: When did we go from agentic loops to graphs?

#2

Post by aleen_oye » Sat, Aug 01, 2026, 5:30 PM UTC

[dead]

Re: Ask HN: When did we go from agentic loops to graphs?

#3

Post by blinded » Sat, Aug 01, 2026, 6:27 PM UTC

Its a fast moving space. Just getting a firm grasp on one pattern should be a starting point for anyone (past basic prompting). For example I just showed a coworker plan and goal style workflows this week.

Its progress, but IMO its not worth workflow jumping until you've got existing purchase on a working flow.

Re: Ask HN: When did we go from agentic loops to graphs?

#4

Post by Johnny_Bonk » Sat, Aug 01, 2026, 10:37 PM UTC

Interesting you post this now because a few months ago I had read about agentic graph RAG or something like that, which was interesting for tracing agentic flows and then just today on Linkedin I came across https://polygres.com/ which now im going to try and excited to see if it has meaningful results for my codebases. To me i've always liked Direct corpus interaction more than rag.

Re: Ask HN: When did we go from agentic loops to graphs?

#5

Post by Yevanchen » Sun, Aug 02, 2026, 5:59 PM UTC

[dead]