ramoz

15 ms·

271.
by ramoz 8mo ago
Love that OP's previous post is from 2024: Rabbit R1 - The Upgraded Replacement for Smart Phones
272.
by ramoz 9mo ago
Not seeing how the sandbox prevents anything really. The point of OpenClaw is to connect out to different systems.
273.
by ramoz 9mo ago
I dont understand these questions/references. It's different because it's a capability baked into the actual tool and maintained by the originators of the tool.
274.
by ramoz 9mo ago
When all of industry is trying to catch up with the features of one coding agent - it may be a signal to just use that one.
275.
by ramoz 9mo ago
I ran a similar operation over summer where I treated vibecoding like a war. I was the general. I had recon (planning), and frontmen/infantry making the changes. Bugs and poor design were the enemy. Planning docs were OPORD, we had sit
276.
by ramoz 9mo ago
This isn't limited to chat bots. There are clearly some developers experiencing it with coding agents. Maybe humans don't carry the cognitive capacity for so much, and so rapid, information-parsing; and we experience vulnerabiliti
277.
by ramoz 9mo ago
This would be considered a failure, or are you saying they don't need to?
278.
by ramoz 9mo ago
I mean, look at Kubernetes though. You have to understand both the application and the infrastructure in order to get the deployment right. Especially in any instance of having to pin the runtime to any type of resource (certain disk writin
279.
by ramoz 9mo ago
Well, I'm a creator and I now have a $4,000 crypto wallet. I did not "pump it" at all. I literally woke up to messages saying, "Hey, you just need to go tie your Twitter account," and I did, and boom. If the busines
280.
by ramoz 9mo ago
https://bags.fm/ - learned of this site this morning, and you can see the creators that are tagged to their royalties. It does seem that some of them are achieving quite high numbers here. They seem to be actively participa
281.
by ramoz 9mo ago
I got bombarded this morning. I woke up and checked my X-account - which I've never tweeted from before (other than replies) - it was being tagged everywhere and I had a ton of messages. I was sent a link to Bags app. And there was som
282.
Show HN: Claude Code Plan Mode Plugin(youtube.com)
3 points by ramoz 9mo ago|0 comments
283.
by ramoz 9mo ago
A much simpler setup can achieve "I want [coding agent messages] on slack"
284.
by ramoz 9mo ago
No, right in the weeds of delegation. I reached out on one channel that you'll see.
285.
by ramoz 9mo ago
This doesn’t solve the problem. The lethal trifecta as defined is not solvable and is misleading in terms of “just cut off a leg”. (Though firewalling is practically a decent bubble wrap solution). But for truly sensitive work, you still ha
286.
by ramoz 9mo ago
I would like to know more. I’m with a startup in this space. Our focus is “verifiable computing” via cryptographic assurances across governance and provenance. That includes signed credentials for capability and intent warrants.
287.
by ramoz 9mo ago
Right. It might be even as complicated as requiring theoretical solutions or advancements of Rice's and Turing's.
288.
by ramoz 9mo ago
That's one thing for sure. I think the bigger problem for me is the rice's theorem/halting problem as it pertains to containment and aspects of instrumental convergence.
289.
by ramoz 9mo ago
One concern nobody likes to talk about is that this might not be a problem that is solvable even with more sophisticated intelligence - at least not through a self-contained capability. Arguably, the risk grows as the AI gets better.
290.
by ramoz 9mo ago
Sandboxes are an overhyped buzzword of 2026. We wanna be able to do meaningful things with agents. Even in remote instances, we want to be able to connect agents to our data. I think there's a lot of over-engineering going there &
291.
by ramoz 9mo ago
We just served Deepseek R1 on this bad boy in CC+TEE (and an integrated signing layer we developed for vLLM). https://pasteboard.co/k1hjwT7pWI6x.png reach out if interested in collab.
292.
by ramoz 9mo ago
Correct, & that is another fun venture in agentic security.
293.
by ramoz 9mo ago
Modern TEE is actually performant for industry needs these days. Over 400,000x gains of zero knowledge proofs and with nominal differences from most raw inference workloads.
294.
by ramoz 9mo ago
Without diving too technically here there is an additional domain of “verifiability” relevant to ai these days. Using cryptographic primitives and hardware root of trust (even GPU trusted execution which NVIDIA now supports for nvlink) you
295.
by ramoz 9mo ago
Recall itself is absolutely ridiculous. And any solution like it is as well. Meanwhile, Anthropic is openly pushing the ability to ingest our entire professional lives into their model which ChatGPT would happily consume as well (they'
296.
by ramoz 9mo ago
yet I was downvoted and while the great HN giant is in newfound agreeance.
297.
by ramoz 9mo ago
yes contain the network boundary or "cut off a leg" as you put it. But it's not a perfect or complete solution when speaking of agents. You can kill outbound, you can kill email, you can kill any type of network sync. Data ca
298.
by ramoz 9mo ago
So sandbox and contain the network the agent operates within. Enterprises have done this in sensitive environments already for their employees. Though, it's important to recognize the amplification of insider threat that exists on any
299.
by ramoz 9mo ago
We're doing so much planning and reviewing with coding agents like Claude Code and OpenCode. I spent a day over break building a better UX for reviewing coding agent plans. Plannotator - Annotate and review coding agent plans visually,
300.
by ramoz 9mo ago
Yea, I'm in a particular health community. A lot of anxious individuals, for good reason, end up posting a lot of nonsense they derived from self-influenced chatgpt conversations. That said, when used as a tool you have power over - Ch

More ›