arthurcolle

5 ms·

31.
by arthurcolle 5mo ago
I found that keeping current context utilization at 18% of total context length was best for minimizing spend, across all models with 400k context length or more
32.
by arthurcolle 5mo ago
did you keep a few of your gmo cultures?
33.
by arthurcolle 5mo ago
Could just be an experiment or something. It's Monday, the week is young
34.
by arthurcolle 6mo ago
As a unit this is funny, Jira points assigned per second (now possible with parallel tool calling AIs)
35.
by arthurcolle 7mo ago
Safari has always rendered HN as the desktop site for me.
36.
by arthurcolle 7mo ago
Writing a book takes like 2-3 years on average. Papers are published everyday. Having a cute two-person "conversational chat" w/ audio works for a lot of people vs. just reading a paper. "No benefit" to you perhaps.
37.
by arthurcolle 7mo ago
I've had a positive experience participating in ApartResearch.com research sprints. I was able to win (with my team) 4th place on the Defense Acceleration hackathon and then most recently on the Technical Governance Challenge, we won 1
38.
by arthurcolle 7mo ago
Is that true?
39.
by arthurcolle 7mo ago
possibly the spending required to do all the severances
40.
by arthurcolle 7mo ago
Is Japan still the largest creditor in the world???
41.
by arthurcolle 7mo ago
$5 = $5 but $5 that I amortize over 7 years might end up being $1.7 maybe if I don't rapidly combust (supply chain risk)
42.
by arthurcolle 7mo ago
Recursive self improvement and Repetitive Strain Injury being the same initialism is really funny to me
43.
by arthurcolle 7mo ago
why don't you ask the model?
44.
by arthurcolle 7mo ago
RIP bash.org Did we ever get a black bar for it?
45.
by arthurcolle 7mo ago
I have used RH to get data for experiments but I heard they insta-ban any API usage to actually execute trades FWIW
46.
by arthurcolle 7mo ago
did RH open up API for trading?
47.
by arthurcolle 7mo ago
Why can't we just introduce a "vouching for" system like lobste.rs
48.
by arthurcolle 7mo ago
Do the end users really care about the models at all, or about the effects that the models can cause?
49.
by arthurcolle 7mo ago
There is a lot of opportunity here for the AI infrastructure layer on top of tier-1 model providers
50.
by arthurcolle 7mo ago
My agents will follow this repo with great interest
51.
Show HN: DSCO agentic CLI with multi-turn tool use and swarms(github.com)
1 points by arthurcolle 7mo ago|0 comments
52.
by arthurcolle 7mo ago
Make this a paid service! This could go viral
53.
by arthurcolle 7mo ago
Growth decoupled from labor costs
54.
by arthurcolle 8mo ago
This doesn't say Axios. Axiom seems to be an exchange.
55.
by arthurcolle 8mo ago
Not a single decent photo of the butterfly robot
56.
by arthurcolle 8mo ago
They mention Mac mini! They have like 3 other desktop lines going, and they mentioned the mini! Also: "In Houston, workers assemble advanced AI servers, including logic boards produced onsite, which are then used in Apple data centers
57.
by arthurcolle 8mo ago
Apple ramping up Mac mini production in Houston to meet demand for Clawbots is wild. When were Mac minis a hot commodity before three weeks ago?
58.
by arthurcolle 8mo ago
I agree with you. I can imagine that people without children that have pets might feel like it's similar, but there's definitely quite a few several hundreds of thousand of years of evolution driving very distinct reactions. Los
59.
by arthurcolle 8mo ago
That's fair. I'm not an evil cloner. I was just curious, one could say, more or less.
60.
by arthurcolle 8mo ago
Thank you for the links, I appreciate this. :)

More ›