Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
omneity
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
19 ms
·
271.
▲
by
omneity
2y ago
Did you procure the servers directly from Dell or through a distribution partner?
272.
▲
Copied Act would make removing AI digital watermarks illegal
(theverge.com)
27 points
by
omneity
2y ago
|
21 comments
273.
▲
by
omneity
2y ago
Thanks for the in-depth explanation! I’ll keep an eye on it :) Fine-grained crash recovery does sound like a great application as well.
274.
▲
by
omneity
2y ago
Very cool concept, thanks for sharing this. Is mandala designed for notebook-style interactive environments only or also when running python scripts more traditionally, in which case could this be integrated in a gitops like environment? (p
275.
▲
by
omneity
2y ago
Another ZIRP casualty?
276.
▲
by
omneity
2y ago
As a long time user of pgvector I'm really hyped up about this. Korvus has the potential to reduce a lot of the repetitive code in projects I work on. You mention pulling models from huggingface for document embedding. Is it possible t
277.
▲
by
omneity
2y ago
The first thing I do on any VS Code fresh install is to switch the sidebar to the right. Pure heresy for many people, I know. But I want my eyes to naturally land on code, not on a file tree.
278.
▲
by
omneity
2y ago
That's just incredible! Thank you so much for sharing :)
279.
▲
by
omneity
2y ago
I did say the index, not the embeddings themselves. The index is a more compact representation of your embeddings collection, and that's what you need in memory. One approach for indexing is to calculate centroids of your embeddings. Y
280.
▲
by
omneity
2y ago
Tried and deployed in production with similar sized collections. You only need enough memory to load the index, definitely not the whole collection. A typical index would most likely fit within a few GBs. And even if you need dozens of GBs
281.
▲
by
omneity
2y ago
> In 2022, production-grade vector databases were relying on in-memory storage This is irking me. pg_vector has existed from before that, doesn't require in-memory storage and can definitely handle vector search for 100m+ documents
282.
▲
by
omneity
2y ago
What is separating this concept from dataflow programming, perhaps with slightly different semantics and interaction patterns? (I reckon my understanding might be limited so I'm genuinely seeking enlightenment)
283.
▲
by
omneity
2y ago
And here I was secretly hoping someone managed to make a minimal MMU-free linux distro to run on ESP32. This is the last I've heard of such an effort: https://www.cnx-software.com/2019/12/03/western-digit
284.
▲
by
omneity
2y ago
My bad for misunderstanding the tone of your comment. I likely wouldn't have gone off in this direction if you asked me what measures I took to check the OTP code in a collaborative/constructive way rather than expressing doubt th
285.
▲
by
omneity
2y ago
Pretty uncharitable interpretation on your end. I am a developer and went to the extent of verifying the content of the QR code and the optional URL for manually adding it. No OTP code in there. Try to be nicer next time.
286.
▲
by
omneity
2y ago
This is simply not true. Just today an app asked me to use Microsoft Authenticator, and scanning the QR code with Google Auth and Authy didn’t work, earning me an “Invalid QR code” error and forcing me to install the app from MS.
287.
▲
by
omneity
2y ago
What makes this an "AI network" or an "AI Superplatform"? Looks like an average Chatgpt UI clone to me.
288.
▲
by
omneity
2y ago
Continuing on the topic. This must be a joke. -- User Your app literally says "Multimodal canvas" by labs.google Gemini 2.27s You are absolutely right! I apologize for the misunderstanding. I am still under development and learnin
289.
▲
by
omneity
2y ago
Tried it out, with the following result. > I can't see or analyze any image. I am a text-based chat assistant and thus I cannot process any image. Pretty hilarious honestly.
290.
▲
by
omneity
2y ago
This sounds awfully similar to actors in Actor Based Programming. Side note, are there other similarly seminal books you can recommend on AI?
291.
▲
by
omneity
2y ago
(another happy paying customer of Fastmail here) I am pleasantly surprised that Fastmail has no AI cruft in it especially that Fastmail is founded by one of the godfathers of modern AI, Jeremy Howard.
292.
▲
by
omneity
2y ago
One workable pathway is to join a team working on AI products as a full-stack engineer. AI practitioners notably are less inclined to do frontend or even backend work, even more so when talking about SWE best practices and moving code to pr
293.
▲
by
omneity
2y ago
I searched for "Supabase" and none of top the results in your demo contained an actual post about supabase. Following the example queries I then tried "What is supabase?" and the results were equally irrelevant. My perso
294.
▲
by
omneity
2y ago
Interesting bit about the gift policy, thank you for sharing. "Limited resources" might be a more appropriate term. Good luck!
295.
▲
by
omneity
2y ago
Pretty whimsical but confusing and borderline creepy, especially the deepfakes and voice cloning without permission. One more thing caught my attention. > all the [...] poverty of us Given all the apparatus, robot dog, iPad per delivery
296.
▲
by
omneity
2y ago
Great work! Is the seed random? No matter how often I refresh it reaches a “dead” state after a few epochs only.
297.
▲
by
omneity
2y ago
Just wanted to mention that Hex Fiend is simply the fastest editor ever (on Mac). I threw gigabytes-sized files at it and it still manages to open them instantly, to the extent that I also use it as a plaintext editor in some extreme cases.
298.
▲
by
omneity
2y ago
This problem extends beyond character encoding. The average joe (and not so average alike) seems to have a hard time to distinguish official channels from non-official scammy ones, even more so when the official channel doesn't exist o
299.
▲
by
omneity
2y ago
In this case isn’t biogas a much better option efficiency-wise?
300.
▲
by
omneity
2y ago
At this feeble level of efficiency we’re probably better off using the energy to split water into hydrogen or producing ammonia. It’s a trade-off ultimately, either we get to use currently deployed ICE systems and feed them with this solar
More ›