ampdepolymerase

Here's a datasheet for an industrial fabric cutter:

http://www.ruk-tech.com/uploads/137f9456.pdf

The bottleneck is the Mastercard/Visa duopoly combined with the modern banking system. At the end of the day no amount of Patreon vs Shopify style business models for adult entertainment work fixes the fundamental payment risk problem. Onlyfans helped amortize the risk somewhat by spreading across multiple payment providers and controlling the UX tightly but even so they still had to change the pricing policy. Mindgeek/CCBill are the Paypal/Stripe of adult entertainment payment processing and it is not an easy field to "disrupt". You can put the best actuaries and machine learning engineers in a room and it would be doubtful if they could come up with a better risk model without fundamentally altering the method of transaction. (Crypto is often touted as a supposed fix for this sort of things but I am not knowledgeable enough about it to comment, somebody else please feel free to chime in.)

Cough, tbray, cough.

Doesn't literally every optimisation and ML class teaches it nowadays?

So this is Hasura for Cassandra?

Not all VCs are made equal and not all startups can take advantage of fundraising. Network is just as important and it could mean the difference between a 50 million dollars company (bootstrapped) versus a 500 million dollars company (multiple rounds of fundraising with brand name VCs). All these are contingent on the fact that your product has the ability to scale. If it is a rather niche area, then no amount of funding will make a difference.

Doesn't matter in practice if you are really trying to re-create Jurrasic Park. A working dinosaur is already useful for monetisation, even if Hollywood's idea of what dinosaurs should be like does not quite match up to reality.

If it is really "less than 3 inches wide", it is most likely a wheel hub motor with a much bigger diameter. 3 inches thick would be a much better description. (This is all conjecture.)

And also in environments with high oxygen concentration. At one point dragonflies were the size of cats (the cause is different from deep sea gigantism).

Considering the success of Grammarly, it is most likely possible to do it with traditional rule-based/experts systems combined with a tiny bit of modern day ML. But there is nothing available out of the box for that.

It is a pity that a similarly specced ThinkPad would be at least a thousand dollars more expensive.

It is easy to make things fast if you cut out 90% of the UI trimmings. No drop shadows, no shaders. No fancy data synchronization and subtle animations on every click and tap. Everything Winform style bog standard UI. And it will be blazing fast. But nobody wants ugly cross platform apps, so here we are.

In other words journalism is great if it tells your preferred version of the truth and narrative. It is being disingenuous to disparage centralised journalism and at the same time complain about internet platforms with reduced gatekeeping.

Compensation in academia is already bad enough. Take away the boosts to the ego and you are going to make the politics even more toxic, and potentially cripple an entire generation.

Calling the ML model should be done via Django Channels or a similar background queue and not directly within the view (request handler) as if the inputs/inference is too big, it will block and cause issues for load handling.

They shut down. Here's the founder: https://twitter.com/robtaylor78

Most software engineers without a strong background in physics or CFD won't be able to reliably explain how an airfoil works either. Having a wrong mental model (equal transit time of air molecules resulting in lift) does not mean airplanes are not useful.

Reminds me of the old reconfigure.io which used the ideas and syntax of Go's CSP and transformed them into async HDL code. Unfortunately the startup has been shuttered.

http://docs.reconfigure.io/

Can you elaborate on the last point?

WebAssembly FaaS cough. I can think of a dozen companies jumping on that particular wagon, Cloudflare being one of the more prominent ones.

That's where ML comes in, to provide the 'intuition' for pruning the search space.

If you are being pragmatic about things, a game engine is probably the best. For 3D, give Godot a try. Otherwise there are plenty of other commercial solutions (Unity, Unreal, there are quite a few). 2D engines are more common than 3D engines, so simply select one that is most compatible with your preferred tech stack. Game engines will require more upfront work for the GUI part of CAD software, but it will make the actual CAD a lot easier as a large amount of constraints and collision physics already come built in. Modern game engines also come with importers for common CAD file formats, so that will make your life a lot easier too. The modern Unity-style game engine with everything and the kitchen sink included is somewhat of a recent phenomenon. Traditionally game engines are small poorly documented libraries and therefore less than helpful unless you were truly desperate. Also, the CAD field moves slowly when it comes to adoption of new technology and most of the incumbents are still using core in-house technology from decades ago. My above assessment is based on the current state of the art for both fields.

Some helpful links:

http://www.publicworksgroup.com/blog/2012/02/will-gaming-eng...

https://www.unrealengine.com/en-US/spotlights/meshmatic-opti...

https://www.researchgate.net/figure/Unity3D-Game-Engine-and-...

Have you seen Boundary Layer Technologies? You should eat their lunch, they certainly could use more competition.

Most of the facts stated are inaccurate, or at least phrased incorrectly in a pop science manner. What you have written are technical subtleties, not the main ideas that undergird biology. Biology is not an absolute hard science, but it serves nobody to emphasize and blow up the details while rejecting the core ideas completely. For example, if I tell a novice programmer that JavaScript is a compiled language like C, that is a very inaccurate statement. The subtleties of JIT interpreters can be clarified when you already have a clear understanding of the differences between an AOT compiler and a high level interpreter. HN is predominantly filled with people of math and computer science background. Those 'facts' in your comment are at best misleading if you do not already have strong grasp of the central dogma.

What about the Soviet Raketa boats?

What are your thoughts on Gravitational's Teleport?

Those aforementioned products are designed for prototyping, for production Atmel chips are fairly competitive.

The last two pairs are non-issues, both have plenty of funding. For the former however, one misstep and you have the FDA/DHS or one of state medical unions breathing down your neck.

For a BOM, 2 cokes is not exactly cheap once you scale it to tens and hundred of thousands. Imagine a SaaS where you are charged for every single database row, that is the reality of hardware unit economics.