srcreigh

7 ms·

1.
by srcreigh 5d ago
> You'd have to test batching of multiple thinking streams in order to try and raise overall tok/s via layer-wise reuse of the streamed weights isn’t this very straightforward to do..? I thought batching for Qwen models is alre
2.
by srcreigh 5d ago
I think the issue is generalization, if you were more specific about which local models aren’t good enough for which tasks compared to which frontier models in your experience, it’d be a lot more informative
3.
by srcreigh 5d ago
If your networking stack and filesystem and who knows what else are exposed to the guest, its not isolation. It’s better than Docker, but it can’t be compared to Firecracker at all. Firecracker actually minimizes the attack surface whereas
4.
by srcreigh 5d ago
This is great as a first look, but the author is not a developer, so we don't yet know whether a dev can be as productive with local models on M5 Mac Studio compared to a 20x subscription plan. I'm also curious about any new low h
5.
by srcreigh 5d ago
Supports direct filesystem access and who knows what else. It is unsuitable for running untrusted code on Linux.
6.
by srcreigh 5d ago
This post explains how a GPT agent broke out of qemu VM. It could not break out of firecracker. https://blog.trailofbits.com/2026/08/26/vms-wont-contain-cyb... Why? Firecracker mounts very few host systems in
7.
by srcreigh 5d ago
I can launch Astra to deploy changes to my homelab via creating Forgejo issues from my phone. That kind of system is pretty hard to set up without kubernetes. Even if you confine yourself to a dev workstation, having 5 agents concurrently b
8.
by srcreigh 5d ago
K8s jobs don’t run in a secure runtime. K8s jobs don’t give you dynamic zero trust permissions scopes. Restoring a harness in 500ms is really fast, much faster than naively creating a new job downloading session and ‘codex resume’ etc.
9.
by srcreigh 5d ago
Tasks are a good scope for zero trust permissions
10.
by srcreigh 5d ago
I have 6 and ended up needing to use my gaming PC for a build server. I think you could get by with 1 computer, but it’ll have to have a pretty decent machine. Between agents running tests, CI, docker image builds, an average $400 mini PC w
11.
by srcreigh 5d ago
It doesn’t provide isolation. It is not even part of the conversation.
12.
by srcreigh 5d ago
I'm not interested in a VM which supports mounting host filesystems in untrusted Kubernetes pods.
13.
by srcreigh 5d ago
So the agent-substrate checks a _ton_ of boxes. Almost all of the things it offers should be table stakes for everywhere we run not only agents but most software. https://github.com/agent-substrate/substrate (For conte
14.
by srcreigh 7d ago
Thanks! No haven’t heard of this. Do you have any experience deploying untrusted workloads with this tool? care to share some more info ?
15.
Secure VMs for Kubernetes: Hardening Kata Containers(srcreigh.ca)
3 points by srcreigh 7d ago|2 comments
16.
by srcreigh 8d ago
I think there's a really important sense in which Godel's argument is not the full story. IIUC, Godel's incompleteness is less about theorems and more about axiomatic systems. Given an axiomatic system, there are statements w
17.
by srcreigh 8d ago
It's impossible for finite number of LLMs to solve all theorems. This would imply that the busy beaver sequence is computable which implies the halting problem is decidable. For any finite program (eg some LLMs), there is a true math t
18.
by srcreigh 8d ago
Not firecracker
19.
by srcreigh 11d ago
Nope, there is no such universal logic. Godel helped show the opposite actually (incompleteness). I think Scott Aaronson's post is very fascinating explanation of this stuff. https://www.scottaaronson.com/papers/bb
20.
by srcreigh 11d ago
The point is that some statements could be provable, but not with today's proof verifier. "Everything that can be proven" is relative. PA can prove some things, ZF more things. In 200 years we could develop more powerful math
21.
by srcreigh 11d ago
The point is that it has limits in what tools it can use by virtue of being a finite program. Those limits imply that there are some math problems that it cannot solve. The LLM can know about this (and frontier LLMs likely do), but that doe
22.
by srcreigh 11d ago
At high enough N, ZFC is independent of BB(N), and in fact any math axiom system has such an N. The LLM itself is finite, the axioms it knows are fixed, there is an N where BB(N) is independent of those axioms, so the LLM cannot solve it.
23.
by srcreigh 11d ago
No, but you hit the nail on the head, that’s the most interesting part. The proof verifier uses fixed math axioms. The busy beaver function at high enough N cannot be proven with those axioms.
24.
by srcreigh 11d ago
> I think it’s broadly unclear whether (3) is true: i.e. whether frontier AI models aren’t generating or can’t generate new mathematical ideas. ... I give basically zero credence to the idea that AIs are incapable of this because of some
25.
by srcreigh 11d ago
What do you think the concrete problem is?
26.
by srcreigh 14d ago
LLMs can help people understand mathematical principles too. The idea that when LLMs produce solutions, people won’t try to understand them and won’t learn from it, is obviously not true. Terry Tao himself spent time digesting and simplifyi
27.
by srcreigh 14d ago
Yeah. This makes sense. Someone who can dedicate a lot of time to scaling a SaaS serving many customers are helped a lot by AI. I believe busy professionals who already have successful businesses won't have the focus to build a one-off
28.
by srcreigh 14d ago
No, this cannot be the actual issue. Most math textbooks have solutions in the back. That didn't wreck peoples ability to learn math, did it? So we are left to speculate about why solutions in appendix are fine, but LLM solving open pr
29.
by srcreigh 14d ago
At least for now, not everybody can make a private SaaS. Doing that requires skills and focus. Even with SOTA models, as of ~today, there are inflection points where the AI would not be able to make the right choice without a knowledgable h
30.
by srcreigh 15d ago
Without pinning providers, you pay let’s say 2x more for agentic coding, since that many input tokens aren’t cached as you bounce between providers.

More ›