Viewing profile — slewis
slewis
HN member- Joined
- Fri, Nov 12, 2010, 11:20 PM UTC
- HN karma
- 1,399
- Public activity
- 302 items
- HN profile
- View on Hacker News ↗
About slewis
@shawnup on X
Founder/CTO @ Weights & Biases
Recent public activity
-
comment
Comment #46640486
Can you try a simpler less programmery version? "are any of my recent blog drafts unpublished and nearly ready to go?"
- story
-
comment
Comment #44176045
OpenAI created a benchmark for this: https://openai.com/index/paperbench/
-
comment
Comment #42854581
It would be really useful to see these evaluated across some of the same evals that the original R1 and deepseek's distills were evaluated on.
-
comment
Comment #42732208
Hey, that's me! Happy to answer any questions about how this works if folks are interested.
-
comment
Comment #42474098
I've spent tons of time evaluating o1-preview on SWEBench-Verified. For one, I speculate OpenAI is using a very basic agent harness to get the results they've published on SWEBench…
-
comment
Comment #41568712
Is it stateful? Like can I do a run, read the results, and then do another run from that point?
-
comment
Comment #41418213
This 100% matches my experience. I like to jokingly call founder mode: "fine-grained multi-level oversight". Others might call it the derogatory "micromanagement". That doesn't mea…
-
comment
Comment #39952465
Creator describing how this works: https://youtu.be/PHQweR1z7pI?si=BpRlWxtxnRmYaeKG
-
comment
Comment #38469745
Overplay one’s hand: spoil one's chance of success through excessive confidence in one's position
-
comment
Comment #38374702
The memorization use case is brilliant. Put your talk track for a presentation in and say “help me memorize this by quizzing me”. Thanks!
-
comment
Comment #37004889
I call this "keep the fingers moving".
-
comment
Comment #36351245
Clicked the youtube link. An hour and 20 minutes later here I am. Thanks for sharing it that was awesome.
-
comment
Comment #36245193
I like the ambition! You can do some amount of iterative and visual data transformation in Weave now. But maybe like 10% of what you can do with Pandas. Pandas is awesome and lingu…
-
comment
Comment #36242520
Hi! I'm Shawn, founder/CTO at Weights & Biases. We've been working on Weave for a couple years now, and it powers core parts of wandb.ai. It's a UI toolkit built for programmers th…
-
comment
Comment #35663029
How dare you! This is a serious place. ^
-
comment
Comment #34906271
Wow! What a joy to click on the comments and find a positive comment at the top. Thanks for writing this and thanks HN upvoters for expressing your gratitude.
-
comment
Comment #34864660
It’s also surprising at first that infinite series can add up to a specific, finite number. If I go halfway to X, halfway again, and so on, where do I end up? All next realities ar…
-
comment
Comment #34815124
Finally! All you ungrateful users who expect your search engines to pore through billions of pages in milliseconds without complaining are going to have to show some humility for o…
-
comment
Comment #34757412
The response here makes me think most commenters don’t have experience with this particular footgun. To clarify: Python can gc your task before it starts, or during its execution i…
-
comment
Comment #33385358
For one thing: if you’re going to make a post like this, put your contact info in your profile!
-
comment
Comment #32607730
nbdev2, which this article is about, is a solution to this problem. It makes notebooks testable, composable, versionable, and more.
-
comment
Comment #32201853
“Everybody already knows that code running on a computer cannot become sentient,” said the temporarily coherent chemical blob.
-
comment
Comment #31988434
Very cool! This might be a naive question. Are your arithmetic patterns equivalent to dependent types, like those found in Idris?
-
comment
Comment #31907829
Yes, I can certainly see how this is confusing. We will work with our legal team to clarify the terms.