yodon

5 ms·

61.
by yodon 5mo ago
So excited to see this - the big advantage of 1.58 bits is there are no multiplications at inference time, so you can run them on radically simpler and cheaper hardware.
62.
by yodon 5mo ago
Fascinating article. I'm genuinely unclear whether the avowedly fictional security team is detailing a real or fictional postmortem.
63.
by yodon 6mo ago
More people read the Atlantic than read books on college admissions. It is possible and often useful to increase the number of informed people even without adding net new information.
64.
by yodon 6mo ago
Wow - as a JetBrains customer I've been waiting for JetBrains to show what a good Agentic IDE should look like. Excited to try it and surprised to find out about it in a random HN comment. EDIT ADDED: bummer - looks like this only work
65.
by yodon 6mo ago
This is so insightfully and powerfully written I had literal chills running down my spine by the end. What a horrible world we live in where the author of great writing like this has to sit and be accused of "being AI slop" simply
66.
Gas Town from Clown Show to v1.0(steve-yegge.medium.com)
2 points by yodon 6mo ago|0 comments
67.
by yodon 6mo ago
Is Bonsai 1 Bit or 1.58 Bit?
68.
by yodon 6mo ago
Can I drag an email directly onto a Kanban or a Todo list, and prioritize it like a task, and then click on the card or task to go directly to the mail message, in the context of its thread?
69.
by yodon 6mo ago
It defaults to being a wrapper around git when it's not custom implemented, and it's recommended that you alias nit as git so the agent can work the way it normally would, just faster and cheaper.
70.
by yodon 6mo ago
Not sure if that's intended as sarcasm. Losing a >$100M asset is far from insignificant.
71.
by yodon 6mo ago
Apologies if I missed this in the article (or in the first article in the series) - what happens if you add two copies of the layer set? Does performance improve over adding one copy of the layer set?
72.
by yodon 6mo ago
If you look at convolutional neural nets used in image processing, it's super common for the first layer or so to learn a family of wavelet basis functions. Later layers then do recognition in wavelet space, without that space ever bei
73.
by yodon 6mo ago
For all its flaws, TSA (at least under previous administrations) did a lot of design thinking work around how to streamline flows through airports, minimize travel stress and conflict, and optimize to minimize traveler complaints while cont
74.
by yodon 6mo ago
A video (attaching the sensor, seeing the visualizations) would help significantly.
75.
by yodon 7mo ago
You may not be familiar with the prevalence of "hormone patches". Absorption through the skin is a common medical delivery method.
76.
by yodon 7mo ago
Pretty sure if this gets any traction at all, it will immediately turn into a mix of Chatroulette and crypto scams. See also Mythic Quest's discussion of TTP.[0] User Generated Content is only your friend if you have the resources to m
77.
by yodon 7mo ago
How does this compare to cloudflare tunnels?
78.
by yodon 7mo ago
It's hard to imagine a video that provides less information. 11 seconds. Pixelated to nothing. Why waste our time with that?
79.
by yodon 7mo ago
>the next step is to combine Archetype 360 with a variation of Holland Codes / RIASEC (vocational interest areas) Shouldn't the next step be actual validation (which looks nothing like asking people what they think about the re
80.
by yodon 7mo ago
I'm a huge fan of spec-kit, and am actively looking for a replacement for it because spec-kit is no longer maintained by the team at GitHub. Codev looks like it has a lot of good similarities to spec-kit, and like it's something I
81.
by yodon 7mo ago
The OP should read up on business strategy, probably starting with Porter's Five Forces. Hint: Money is a nice thing to have but it is very definitely not a moat. If a company is making above market returns and the only thing stopping
82.
by yodon 7mo ago
<checks stock market activity>
83.
by yodon 7mo ago
When claude or codex does something other than what you want, instead of getting mad at it, ask it what it saw in your prompt that led it to do what it did, and how should you have prompted it to achieve what you wanted. This process tends
84.
by yodon 8mo ago
When you get a chance to work on your login flow, I recommend giving users an opportunity to request the key rather than automatically showing it once only on the first screen. I created the account from my phone, and don't have access
85.
by yodon 8mo ago
The first screen of your signup flow asks for "organization" - is that used as a username or as an organization name or both (I can't tell what if anything will be on the next screen) If your registration process is eventuall
86.
by yodon 8mo ago
For the moment, researching multi-agent orchestration. At first glance, your work looks among the best in class of published work I've seen. Particularly interested to understand the memory/communication/search model you&#x27
87.
by yodon 8mo ago
Can you add a license.txt file so we know we have permission to run this (eg MIT and GPL V3 are very different)
88.
by yodon 8mo ago
If you could actually detect AI content with high accuracy, you would sell it as a service and print money, but you can't, so you force all the rest of us to wade through posts like yours, claiming to tell the rest of us what is and is
89.
by yodon 8mo ago
With all the Microsoft haters in this thread, I'm surprised no one has pointed out this is just another example of Microsoft copying Apple. Satya's speech is essentially the same one Steve Jobs used to give to new VP's at App
90.
by yodon 8mo ago
Providing some sample screenshot+output pairs would be helpful and probably singificantly increase the odds users interact with your product.

More ›