randall

6 ms·

1.
by randall 2mo ago
+1 to this.
2.
by randall 2mo ago
wow! i’ve been using deno for a long time, and one of my fav features is compiling a binary. i didn’t see anything about that, but might have missed something… do you all plan to support this?
3.
by randall 5mo ago
this isn't strictly true. not that it thinks, but it can reason about the tokens that led to the outcome.
4.
by randall 5mo ago
this is "you ship your org chart" not ai. https://en.wikipedia.org/wiki/Conway%27s_law
5.
by randall 6mo ago
Does anyone have a link to the "read more"?
6.
by randall 6mo ago
exactly. They probably have unsustainable margins on accident.
7.
by randall 6mo ago
sweet. I've been out of the game for a bit but it's great to hear.
8.
by randall 6mo ago
This is the least informed take i've ever seen. I worked at fb, and I'm 100% certain we sponsored VLC and OBS at the time. It would be strange if we didn't sponsor FFMPEG, but regardless (as the article says) we definitely go
9.
by randall 7mo ago
even marvel villain thanos was given a plausible motivation. i don’t think if you actually knew him as a person you’d think this was a reasonable perception.
10.
by randall 7mo ago
same. this is about losing a negotiation and saving face / exacting revenge.
11.
by randall 7mo ago
i don’t get it. what’s his motive in your view? he literally has no shares in openai.
12.
by randall 7mo ago
I wanted the AOL one to say "Welcome" before "you've got mail!" lol
13.
by randall 7mo ago
The last 3 years of LLM progress, to me, feel like 1994-1998.
14.
by randall 8mo ago
this is like irl cryptographic signatures for content lol
15.
by randall 8mo ago
thanks for the PR! :)
16.
by randall 8mo ago
yaaaaa exactly. You're totally on the same wavelength as me. Let's be friends lol
17.
by randall 8mo ago
once your team comes to a consensus on what PII is, you can roughly guarantee it... especially as models improve.
18.
by randall 8mo ago
So two things. 1/ crypto signing is totally the right way to think about this. 2/ I'm limiting prompt injection by using chain of command: https://model-spec.openai.com/2025-12-18.html#chain_of_comma... we ha
19.
by randall 8mo ago
i have a huge theory here that idk when we’ll implement but it has to do with “quorums” and other stuff. hard to explain… we’ll keep going.
20.
by randall 8mo ago
right now yeah we’re just dropping context… sub agents are short lived. thinking about ways to deal with that but we haven’t yet done it.
21.
by randall 8mo ago
yeah the way to do that stuff is through zod schemas… input and output schemas. you can set up really complex validation. thanks for checking it out!!
22.
by randall 8mo ago
i have a slightly different but related take. the models actually are getting smarter, and now the challenge becomes successfully communicating intent with them instead of simply getting them to do anything remotely useful. Gambit hopefully
23.
by randall 8mo ago
So I look at something like Mastra (or LangChain) as agent orchestration, where you do computing tasks to line up things for an LLM to execute against. I look at Gambit as more of an "agent harness", meaning you're building a
24.
by randall 8mo ago
thx, i appreciate it, believe it or not. :)
25.
by randall 8mo ago
Thx! Happy to help if you need it. :)
26.
by randall 8mo ago
thx!
27.
by randall 8mo ago
omg thank you so much. We're working on the file system stuff, that's an easier lift for us than the initial work, so we wanted to start with the big stuff and work backward. Claude Code and Codex are obviously really great at tha
28.
Show HN: Gambit, an open-source agent harness for building reliable AI agents(github.com)
91 points by randall 8mo ago|27 comments
29.
by randall 8mo ago
I think that works out tremendously well for Nintendo, especially when you look at the Wii-U vs the Switch. I shot a video at CNET in probably 2011 which was a single touchscreen display (i think it was the APX 2500 prototype iirc?) and it
30.
by randall 8mo ago
Nope, apple container just runs a lot more efficiently on apple silicon macs than docker.

More ›