polynomial

5 ms·

31.
by polynomial 4mo ago
You don't know that No Feigning Surprise is actually from an xkcd comic, before it was a wizardzines post? U+1F632
32.
by polynomial 4mo ago
Honestly, if you compare Tim's pre-GPT writing style, to his current stylometrics, you'll see a remarkably delta. It's as if he suddenly decided to change his established writing style and voice right when LLMs become widely
33.
Treating Agent Reasoning as a Span(forestmars.substack.com)
1 points by polynomial 4mo ago|2 comments
34.
by polynomial 4mo ago
Reasoning chains as first-class OTel spans, which is what makes autonomous agent production deployments auditable (rather than indefensible.)
35.
by polynomial 4mo ago
Less MTTF and more MTTR
36.
by polynomial 5mo ago
This is horrifying on so many levels even if you don't already know how the story ends.
37.
by polynomial 5mo ago
Calling it "Act 2" certainly isn't helping with investors.
38.
by polynomial 5mo ago
A 67 year old woman living out of her car? JFC.
39.
by polynomial 5mo ago
Very obviously LLM written to anyone who's spent any time using them.
40.
by polynomial 5mo ago
Not everybody isn't worried: https://ctolunchnyc.substack.com/p/cto-lunch-nyc-spring-2026...
41.
by polynomial 6mo ago
"The gap is larger than most people realize" is a a dead giveaway.
42.
by polynomial 7mo ago
Fully agree. If you don't change your approach but just use CLI "intead of" MCP, you'll end up with a new spin on the same problems. The guardrails MCP provides (identity, entitlement, multi-principal trust boundaries) s
43.
by polynomial 7mo ago
This is the right framing. The chain policy problem is what happens when you ask the registry to be the entitlement layer. Here's a longer piece on why the trust boundary has to live at the runtime level, not the interface level, and w
44.
by polynomial 7mo ago
Understandable as always a proxy for predictable.
45.
by polynomial 7mo ago
This is the under acknowledged "secret" of this reconfiguration. It's like the Bill Joy point about mediocre technology taken to the next level.
46.
by polynomial 7mo ago
Robots making fun of us complaining about them.
47.
by polynomial 7mo ago
Well at least you know who to fire
48.
by polynomial 7mo ago
"Petri dish rewrites React in Rust"
49.
by polynomial 7mo ago
Tragically this reference is all but lost generationally.
50.
by polynomial 7mo ago
Plastic cutlery is a dead-on perfect analogy.
51.
by polynomial 7mo ago
TBF those RSUs come in handy on cold winter nights when you need tinder.
52.
by polynomial 7mo ago
I've hired that guy and he was a ticket grinder, well-respected by the team.
53.
by polynomial 7mo ago
Classic LLM construction. 5 sentence paragraph. First sentence is parataxis claim. Followed by 3 examples in sentence fragments, missing verbs, that familiar cadence. Then the final sentence, in this case also missing a verb. Pure AI slop.
54.
by polynomial 8mo ago
> The whole thing reads as egotistical, self-absorbed, and an absolute refusal to accept any blame or perform any self reflection. So, modern subjectivity. Got it. /s
55.
Why OpenClaw's observability model can't keep pace with its own scope(ctolunchnyc.substack.com)
4 points by polynomial 8mo ago|1 comments
56.
by polynomial 8mo ago
A $200 Rabbit r1 running OpenClaw for 36 hours showed apps are "really just" just markdown files. But as the agent runtime grew the engine's 4 tool minimalism to multi-agent clusters, the map peeled away from the territory. F
57.
by polynomial 8mo ago
That's the problem with sarcasm...
58.
by polynomial 8mo ago
You need to publish this tbh
59.
by polynomial 8mo ago
Luckily we're having a record cold winter and your setup can double as a personal space heater.
60.
by polynomial 8mo ago
Depending on what your usage requirements are, Mac Minis running UMA over RDMA is becoming a feasible option. At roughly 1/10 of the cost you're getting much much more than 1/10 the performance. (YMMV) https://buil

More ›