frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

Open in hackernews

Ask HN: Anyone using knowledge graphs for LLM agent memory/context management?

9•mbbah•21h ago
I’m building infrastructure for LLM agents and copilots that need to reason and operate over time—not just in single prompts.

One core challenge I keep hitting: managing evolving memory and context. RAG works for retrieval, and scratchpads are fine for short-term reasoning—but once agents need to maintain structured knowledge, track state, or coordinate multi-step tasks, things get messy fast; the context becomes less and less interpretable.

I’m experimenting with a shared memory layer built on a knowledge graph:

  - Agents can ingest structured/unstructured data into it

  - Memory updates dynamically as agents act

  - Devs can observe, query, and refine the graph.

  - It supports high-level task modeling and dependency tracking (pre/postconditions)
My questions: - Are you building agents that need persistent memory or task context?

  - Have you tried structured memory (graphs, JSON stores, etc.) or stuck with embeddings/scratchpads?

  - Would something like a graph-based memory actually help, or is it overkill for most real-world use?
I’m in the thick of validating this idea and would love to hear what’s working (or breaking) for others building with LLMs today.

Thanks in advance HNers!

Comments

frenchmajesty•18h ago
Funny you should ask I just ended up here googling "graph memory LLM"

So yea I'm very much looking into it. I want my personal agent to grow to know me over time and my life is not bunch of disparate points spread out across a vector space. Rather It's millions of nodes and edges that connects key things. Who my parents were, where I grew up, what I like to do for fun and how it ties into my personality and strengths, etc...

To have this represented in a graph which a model can then explore would allow it to make implicit connections much easier than attempting the same with embeddings.

Ask HN: What are good high-information density UIs (screenshots, apps, sites)?

492•troupo•2d ago•359 comments

Ask HN: RAG or shared memory for task planning across physical agents?

10•mbbah•21h ago•1 comments

Ask HN: How much better are AI IDEs vs. copy pasting into chat apps?

134•lopatin•2d ago•132 comments

Ask HN: Escaping a Low-Paying Nepali IT Job and Ineffective Learning Cycle

7•shivajikobardan•11h ago•2 comments

Ask HN: What would you do with the #manga chat channel in Libera IRC network?

4•babuloseo•14h ago•1 comments

Ask HN: Is there a service that offers Common Crawl as an API?

6•georgehill•15h ago•2 comments

Ask HN: Anyone using knowledge graphs for LLM agent memory/context management?

9•mbbah•21h ago•1 comments

Blazeio.SharpEvent: A Python Async Primitive That Scales to 1M Waiters with O(1)

6•anonyxbiz•1d ago•0 comments

Ask HN: AI-Filtering Browser Extension?

7•v-yanakiev•1d ago•3 comments

AI Summarizer: Summarize Web, YouTube and PDFs in Seconds–Free

10•huizhu•2d ago•2 comments

Ask HN: How to get good at marketing your product and SEO?

4•flashblaze•1d ago•4 comments

OSUniverse: Building a Better OSWorld

5•mountainriver•2d ago•0 comments

Ask HN: How could vibe coding show the code at a high level to non-programmers?

6•amichail•1d ago•8 comments

Ask HN: Hackathons feel fake now

210•sepidy•5d ago•128 comments

Ask HN: Are you using AI coding assistance?

8•cloudking•1d ago•13 comments

Ask HN: Nvidia GeForce RTX 5060 arrives May 19 at $299 revive PC builds?

10•byte-bolter•2d ago•11 comments

Ask HN: How do you obtain software development contracts?

30•codingclaws•3d ago•17 comments

Getting tired of Helm – any better way to handle deployments in Kubernetes?

22•DeborahEmeni_•4d ago•21 comments

Ask HN: Did Aliexpress stop shipping to US?

27•olalonde•4d ago•17 comments

Why do websites prevent pasting via onpaste="return false;"

5•gleenn•1d ago•3 comments

Ask HN: Help us validate our idea of an administrative app for small businesses

3•Kuyawa•1d ago•1 comments

Ask HN: How are you managing LLM inference at the edge?

7•gray_amps•2d ago•1 comments

We built an AI-powered voice tool to boost sales

2•Artjoker•2d ago•1 comments

Ask HN: Why is the sender chat box always on the right?

5•bdhe•2d ago•8 comments

Is a Smaller Internet Better?

3•sawyersweet•2d ago•0 comments

Ask HN: What's the best framework for building Mac/Windows desktop apps in 2025?

7•anoojb•19h ago•6 comments

Ask HN: Which Firefox add-ons are you using in 2025?

6•vintageclothldn•2d ago•11 comments

Ask HN: Have you used Claude Code? Is it any good?

8•mbm•3d ago•9 comments

Ask HN: Privacy concerns when using AI assistants for coding?

6•Kholin•1d ago•5 comments

Ask HN: Has anyone managed to pass Meta's Access Verification?

24•hipgrave•5d ago•11 comments