Can you be more specific?

If you take every long-refuted criticism of Bitcoin along with cherrypicked examples of cringe celebrities endorsing it, and you combine that with misinformed neo-Marxism and an unquestioning acceptance of government officials, you get this article.

If your average sports writer actually wanted to understand the space, they could read something with real data: https://nydig.com/wp-content/uploads/2021/09/NYDIG-Bitcoin-N...

> If this is true then the goal is to create a mining algorithm that focuses on being as wasteful as possible.

This wouldn't be profitable for miners, so the economics of the coin would collapse. Bitcoin is designed to balance these incentives.

> This is simply not true. There is no such thing as excess power, whatever is not used should be stored or sold to other countries.

You don't understand power grids. Electricity is diminished the further it travels over transmission lines. Only fossil fuels like oil can be "stored and sold to other countries", renewables don't work like this. Excess power generated by renewables, and even by some non-renewables, is frequently wasted because (at certain times) it's either not economically feasible to transport it to population centers, or the grid is at full capacity. You can't pump that excess power into the grid or you will break it.

> Excess energy production has always been a thing

Sure, but this is a new solution.

> and there are mitigation efforts in place much better than "lets put a server farm that wastes energy"

Why do you think current plans are better? Bitcoin mining gives a financial incentive to produce renewables as a direct result of the technical details of power grids. You're being awfully dismissive for someone who knows so little.

> I see modelling is almost always only open to phds/masters.

I think this varies pretty widely based on employer, e.g. if you are at a smaller company (and you show interest and have the necessary skills) then you're much more likely to be able to contribute on the modeling side. It's easier to get there if you have an advanced degree, but definitely not necessary.

That being said, IMHO book lists like this aren't very useful because there's no incentive to keep them short and realistic. Reading seminal papers and implementing them is a different learning philosophy, maybe, but lists like this are probably more feasible to complete: https://dennybritz.com/blog/deep-learning-most-important-ide...

Good find, and IMHO this should probably be the default integration in their docs; the first tutorial listed here is for the Payment Intents API, which is a more complicated version of the Charges API: https://stripe.com/docs/payments

The Dream Machine is very long but IMO very good and worth the effort. Part-biography, part-history (going all the way back to the early days e.g. Turing), and extremely thorough!

I work at a company that uses word embeddings to generate predictions that customers actually pay for. There are lots of other things going on under the hood (e.g. neural networks in which the embeddings are the first layer, non-ML software architecture to make the UX something worth paying for) but to answer your question, embeddings are definitely generating real value.

> most people just want to drop a dataset and pick the type of model (say, multiclass image classification) and leave the rest for machines to optimize.

I think the disconnect here is that you can reuse existing architectures and get state-of-the-art performance without running something like AutoML. It's not clear that creating a bespoke architecture for your specific problem is always better, let alone always a good use of your resources.

Tempo Automation | Full-Stack Software Engineer | SF | Full-Time | ONSITE | https://www.tempoautomation.com

Tempo operates a factory in SF to produce printed circuit boards extremely quickly - in doing so, we give our customers the physical equivalent of a fast compiler, enabling them to turn their designs into real products significantly faster than they could before. Our customers build rockets, drones, medical devices, and much more - working at Tempo means you can accelerate all of these industries at once.

Providing a system to produce circuit boards quickly and correctly means we're both a software shop and an advanced manufacturer. Our software team is still pretty small (6 engineers) which means any new hires will still have an outsize impact on the direction of our technology, and on the company as a whole.

We love hiring strong generalists who take initiative and act as part-time PMs when they're working on projects.

Some projects you might work on:

  - Offering real-time, automated manufacturing feedback based on certain parameters of our customers' designs. Think linting for hardware.

  - Optimizing our factory processes with machine learning and scheduling algorithms.

  - Encoding the three-dimensional structure of each order's components (e.g. resistors, capacitors) for our assembly robots, in order to speed up assembly and ensure minimal manufacturing defects.

  - Optimizing our automated optical inspection machine to make it faster than any in the industry.

We're also excited to be moving into a MUCH larger factory (>4x the floor space) in SoMa in a couple months, which means you'll see that space evolve from nothing into an enormous, high-tech manufacturing complex.

If you're interested, please apply at https://www.tempoautomation.com/careers or send a note to george [at] tempoautomation [dot] com.

Tempo Automation | Senior Software Engineer (Full Stack) | San Francisco, CA | ONSITE https://tempoautomation.com

Tempo is accelerating the pace of electronics development. We make printed circuit boards quickly and reliably, to ensure electrical engineers have a much tighter feedback loop when they're iterating on their designs. By combining quick-turn manufacturing with automatic CAD analysis + instant design feedback, we're trying to give electrical engineers access to extremely high-quality dev tools, similar to what SWE's are already used to.

We run a factory in SF, so working as a full-stack SWE (we currently have 4) means you get full ownership over projects as diverse as automating factory equipment, optimizing logistics, analyzing CAD files, and delivering beautiful customer-facing features on our website.

Our customers make this job awesome: we're helping rocket companies launch satellites into space, AI companies produce custom hardware, healthtech companies advance robotic surgery, and much more.

Stack: ES6+React, Python+Django, Postgres, and AWS.

Interview process: Non-technical phone screen, code challenge, technical phone screen, onsite.

--

https://jobs.lever.co/tempoautomation/5c1f5e85-a195-4fc3-ac8...

I think the point is that he could use his immense power and unique platform to win regardless of what you or I might want.

And to put it in context, powerful businessmen with no prior gov't experience are now emboldened to run for the highest office (http://www.reuters.com/article/us-usa-election-dimon-idUSKBN...) which, as an aside, doesn't require putting your assets in a blind trust.