amorroxic

Curious on what backs this assertion. As a counterpoint we’ve been running 200+ models in production for more than 5 years - language models, embedding, classifiers, low tens to hundred M params. Traffic in the order of 1-2M requests/day and everything is enabled by onnx with some cgo (or Rust) plumbing on top. What’s your SLA?

Thanks a lot for this. Also one question in case anyone could shed a bit of light: my understanding is that setting temperature=0, top_p=1 would cause deterministic output (identical output given identical input). For sure it won’t prevent factually wrong replies/hallucination, only maintains generation consistency (eq. classification tasks). Is this universally correct or is it dependent on model used? (or downright wrong understanding of course?)

A real-time video processor in hardware:

- static (mp4) or live (hdmi) input, hdmi output

- a collection of effects (various distortions, color glitches, demoscene-like) applied over incoming video stream, maximum of 4 effects stackable on top of one another

- midi controllers support (controlling actions/params of effects via CC)

- modulation of effects params via LFO and audio events (bpm, kick, tonal detection)

- loading of .glsl shaders (eq. shadertoy.com)

- dynamic input resolution, output either 360p (rpi 4/5) or upscaled to 720p / 1080p (networks like SRGAN over Hailo / RPI or RK3588 with a Radxa 5B SBC).

Given a 2nd screen (timeline editor) would love to evolve this to something like a hardware editor, somewhat in the line of DAWs in the audio world. Most things are working with biggest challenge now being building a control surface (buttons, rotaries + associated oleds, etc) and attempting laying it all on a PCB, a process I don't know much about. If there's interest welcome comments and could elaborate more.

Maybe this could be useful? https://www.playonlinux.com/en/ (able to run Photoshop myself)

first $100: pdf photography courses with weekly chapters

first $1000+: built a shop selling silk scarves (100-odd) I collected myself from Laos/Cambodia/Thailand to balance trip budget (25yo back then, mad travelling). Sold all scarves, closed shop.

the null concept being referred to as “the billion dollars mistake” - until discovering optionals and gradually opening up to FP, category theory, etc

This. Same scenario here, fondly remember 1080ti, that was the workhorse for us alongside cloud training (startup with ~100 nlp prod models, ner, siamese, count models, etc). ULMFiT and language transfer was the moment when upgrades felt necessary though 2080ti’s vram unfortunately stayed the same (never had access to Titans for instance so perspective may be limited).

and unfortunately social makes it possible for one to yell “Fire” in all crowded places at the same time.

it’s still there unfortunately, no close button either.

Signed it too :) Can’t remember a particular reason though in our case (former cofounder) we were discovering the actor model, immutability, streams,.. tbh remember a great deal of fun programming during those days, insane schedules and whatnot. Though truth be told our most productive systems are Go these days (sold, stack still holding).

Three strikes and you’re banned forever, better choose the moment well.

Thank you for the perspective here. A wish there was a way to bookmark comments on HN..

Agree, this being a cyber-fraud company I believe one could keep a team busy for a long time focusing on building secure systems rather than operating them to the point where no-one gets to see what's actually being stored within. The paper trail though I'm still puzzled about, he couldn't have acted alone.

In the US maybe, gig economy is global though.

Man, don't beat yourself up. Americans are great people and not that bad of a society now. (someone in a WE society). Hang in there, we'll bicker after :)

Genuinely interested in good numbers to compare then, any chance to elaborate?

You could have a look at ArangoDB (https://www.arangodb.com), a multi-model database - it may cover your use cases. Hopefully linking is ok in comments, will remove if its breaks community rules - no affiliation etc..

What's the point of a random comment on the internet when there are far more efficient ways to spend one's time?

thanks for this perspective.

Same, Focus is great. Also a very warm mention for what I think became one of my favourite features while browsing - one single tab.

Involved in development with pretty insane schedules, 14-18+ hrs/day pretty often and weird as it sounds enjoying my activity immensely. Marijuana keeps me focused at times without a single Alt+Tab for most of the day. Far from ideal or healthy and I wouldn't recommend this to anyone however truth be told there's also a perennial race against time, if I'd be to stop the company would probably grind to a halt.

Never talked about this with anyone and would never even dare interacting with people in public in an altered state of mind but luckily we're a decentralized team. I'm also conscious about the bad part of this being gradual social retreat/isolation and inability of engaging with people at chit-chat/informal level. In my case though satisfaction from seeing an algorithm flying is just too great - but I'm also lucky of having a really supportive (non-smoker) partner, life without this aspect would most probably look much much different.

Sentence similarity were my explorations with WMD too, reached a setup in Keras with a siamese configuration, Wasserstein + KL loss (have a known vocabulary and feeding both word vector sequences as well as their LDA distributions as input). Post training cosine distance between encodings of such sequences look pretty decent - with one issue I've spotted though: WMD really seems to like about the same number of valid tokens in both sentences which is not how real world looks like - eager to see results of EM distance between image feature vectors, cheers.

So look. 3 hours after you wrote this (now, that is) you probably forgot your comment (whether a random HN project becomes a business or not). Here's my take for an answer - it doesn't matter :)

It probably touched base for everyone linking such mirrors with mental health points - fair play, I know I'll think about this tomorrow regardless of its far-fetched potential biz-prop. Cheers!

I like it, really! I believe this to be a good use-case for decentralization as opposed to anonimity - if paired with a good verification mechanism (so leaving behind the crypto-anonimity mantra) this would also guarantee integrity of timelines (tampering/algorithmic curation of news and what not). Good luck to you Sir!

> Or Google and Facebook will be held to a higher standard of compliance.

Than whom else? VK? Baidu? ICQ?

This would be a valid point or topic of debate should the subject be an actual law (the 70 years copyright law for instance?)

What we're dealing with here is loopholes and nothing more under various incarnations - take patents in the US or the lax tax policies of Malta or Isle of Man as an example - and you'll get a sense on why a bit of cleaning up is sorely needed.

European here and I feel this is a bit shortsighted. Not really sure the sentiment is to target US companies (if for nothing else ours are quite open/interconnected economies and there's lots of fondness for US services - check both market share as well as revenues) but really.. Facebook's tax of 5k/year in the UK, the double irish with a dutch sandwich, the backroom deals from the 80s with the Irish govt.. these are global practices which truly devastate everyone in the long run. I actually trust bureaucrats to step in and audit - and subjectively think they very often do a decent job of explaining rationales given europeans' known tendency of bickering on pretty much everything.

On the same page. Have gone through endless flavors of Linux, Win and OSX by simply copying the emails folder from system to system for close to 20 years now - and never lost one email. To say I appreciate this piece of software would be an understatement.

Props to you guys, outstanding product really.

ArangoDB is an incredible database with some very shy marketing imo - edge indexing over json documents is too good of a sweet spot.

Not sure if still actual, however among headaches I remember race conditions in the node driver(rocksdb engine) and maybe less straightforward clustering (compared to Couchbase/Cassandra/etc at least).