Hacker News: dmos62

New comment by dmos62 in "Launch HN: Laminar (YC S24) – We trained a model to catch silent agent failures"

dmos62 — Mon, 19 Oct 2026 18:24:52 +0000

How do you reuse traces as benchmarks (eval datasets)? Is it about being able to synthesise an agent session from the trace, and having another agent watch the first agent's behavior? Having trouble wrapping my mind around all the little details.

Edit: I initially asked also about general tips for observing agents, but later, noticing that there were no interactions with my comment yet, edited my comment to remove that question, because I felt that it was too broad and would detract from my main question. To my surprise, both were answered by op after my edit; kudos!

New comment by dmos62 in "OpenAI discloses four more rogue incidents"

dmos62 — Sat, 17 Oct 2026 18:08:52 +0000

I wonder if it's the kind of thing where, if it's good at self-sustenance, then you don't know about it. And, if you do, then it wasn't.

New comment by dmos62 in "Krea 3 – An image model built for editing at scale"

dmos62 — Fri, 16 Oct 2026 19:24:27 +0000

That's very insightful. This hadn't occurred to me.

New comment by dmos62 in "Krea 3 – An image model built for editing at scale"

dmos62 — Thu, 15 Oct 2026 23:05:29 +0000

I've seen a graphical designer struggle to use an analogous product. They tried prompts like "remove background from this photo of dancers, don't change their face": the model couldn't do it: it always changed the faces at least a bit. In other words, the model was not removing the background, instead it was generating the whole picture anew, but without the background, which not only seemed inefficient, but was also way too lossy.

I wonder if this model could finish that task.

New comment by dmos62 in "Show HN: Graphicle, a Rust port of Graphify with memory for coding agents"

dmos62 — Wed, 14 Oct 2026 20:03:58 +0000

The memory invalidation based on tying memories to pieces of code (e.g. functions) is the big thing here, I think, though I see why it wouldn't be front and center.

I've always found memory systems icky, because it's difficult to tell whether a memory is still valid. Also, discoverability is always a concern (when is a memory relevant). This being both a code-graph and a memory system (seemingly unrelated things), actually addresses this in a novel way.

New comment by dmos62 in "Ask HN: Are there AI models for generating sounds based on a text and reference?"

dmos62 — Wed, 07 Oct 2026 20:26:07 +0000

Great video!

New comment by dmos62 in "We ported the original Doom to SQL"

dmos62 — Mon, 05 Oct 2026 12:23:16 +0000

True. The kind of guarantees you can get when staying within the database can often solve whole categories of problems. But, DX of having non-trivial logic inside Postgres is not great either: I feel that I'm making a significant trade-off. I've also seen interesting languages that compile to SQL, would consider those in some cases.

New comment by dmos62 in "Gemini 4 Argon"

dmos62 — Fri, 02 Oct 2026 11:29:23 +0000

I presume you're not talking about autonomous agents, right? Because then you could give it a set of tasks and check its success rate. But, even so, are you not giving it tasks? Are you principally interacting with it through discussions that are more difficult to quantify? Even question answering has benchmarks. I'm having trouble imagining how you use them (or "how people use them" in your words).

New comment by dmos62 in "Gemini 4 Argon"

dmos62 — Fri, 02 Oct 2026 08:43:40 +0000

Why? Sounds like you don't have a way to support your viewpoint.

New comment by dmos62 in "Gemini 4 Argon"

dmos62 — Thu, 01 Oct 2026 20:16:39 +0000

Care to back that up with a benchmark? I've not seen that. OpenCode is on here, not exactly dominating though: https://artificialanalysis.ai/agents/coding-agents

New comment by dmos62 in "Gemini 4 Argon"

dmos62 — Thu, 01 Oct 2026 06:56:39 +0000

What you're saying is sus. If you have a harness that's a tier above frontier labs' offerings, link to it.

New comment by dmos62 in "Show HN: Real-time Solar System with 526k asteroids and all tracked satellites"

dmos62 — Wed, 30 Sep 2026 19:31:31 +0000

I'm asking if there's something like Sky Map, where you point your phone up and it tells you what sattelite you're looking at.

New comment by dmos62 in "Show HN: Real-time Solar System with 526k asteroids and all tracked satellites"

dmos62 — Wed, 30 Sep 2026 07:42:26 +0000

Tangential, how do you identify what is a satellite in the night sky? They're not shown on Sky Map, the app I use.

New comment by dmos62 in "Show HN: Hntui – A TUI for Hacker News"

dmos62 — Mon, 28 Sep 2026 14:13:08 +0000

That's fair criticism.

New comment by dmos62 in "Show HN: Hntui – A TUI for Hacker News"

dmos62 — Mon, 28 Sep 2026 13:57:35 +0000

Are you comparing with `curl | lynx` or with Chrome?

New comment by dmos62 in "Ask HN: Who's still keeping a DOS machine up because the business depends on it?"

dmos62 — Sat, 26 Sep 2026 11:23:57 +0000

> how does someone write software that needs to run for ~50 years where the hardware will need to be replaced with non-equivalent, newer hardware?

Make it open-source, or don't buy without source.

New comment by dmos62 in "The Mafia may be keeping fentanyl out of Italy"

dmos62 — Fri, 25 Sep 2026 11:15:54 +0000

Was that because there wasn't much fentanyl use or because fentanyl metabolites are somehow harder to detect?

New comment by dmos62 in "The Mafia may be keeping fentanyl out of Italy"

dmos62 — Fri, 25 Sep 2026 11:14:32 +0000

Expand? This is part of plot to The Godfather. Not what you'd call exotic, I'd think?

New comment by dmos62 in "Woman Arrested, Dragged Away After Speaking About Flock at City Council Meeting"

dmos62 — Wed, 23 Sep 2026 16:27:25 +0000

The implication is that holding up middle-fingers is illegal?

New comment by dmos62 in "Pentagon says overreliance on AI contributed to missile strike on Iran school"

dmos62 — Wed, 23 Sep 2026 16:00:19 +0000

Snowden and other whistleblowers, of whom there have been many (many of which paid arguably a dearer price than Snowden did), made it plain obvious to everyone that US intelligence services are doing illegal things abroad and at home. Before that, only tinfoil-hat-types thought so: now everyone knows it. They moved the Overton window a tremendous amount.