venantius

It has been pointed out previously that BART's infrastructure was totally reasonable...for the original level of growth the city assumed it would face. Meanwhile, SF's actual population growth has exceeded that to a considerable degree.

Griffin | Full-stack or Backend Engineer | Full-Time | REMOTE | London, UK | https://griffin.sh

We're rapidly moving towards a world in which people get most of their financial services from non-bank fintechs than they do from banks. The problem is, most of those companies need to partner with a bank to actually hold their cash. We're building a purpose-built bank to do exactly that - power the global fintech revolution.

We're applying for full regulatory permissions and building a brand new platform bank, launching in the UK first. We've been working on this quietly for about 18 months and are now looking to hire our first engineer.

Tech stack: Clojure / Postgres / Kafka / AWS / Kubernetes

Email jobs+hn@griffin.sh

The 1% refers to a specific sub-segment of the overall capital requirements. This would be on top of any normal Basel III requirements and is typically marked as a "discretionary" capital buffer to be set by the state in question.

EDIT: Ah, nuts, meant to reply to grandparent rather than parent. Oh well.

It's possible but pretty unlikely. The capital buffers being introduced are about 1% of Danish banks' balance sheets, and it's likely that the capital buffer for this was previously non-zero (e.g. likely 0.25% or 0.5%). So an increase of 0.5% or 0.75%.

And it's worth keeping in mind that a lot of banks, for governance reasons, will already have been holding in excess of the regulatory requirement. So it's not as if most of the banks will have to come up with this money suddenly - it'll already be there and available.

That's not to say that it's not a large amount of money in absolute terms, but relatively speaking if they were to reduce their loan origination for the year they have to come up with the money be something like 2-3% they'd be totally fine. That's not exactly going to kill the economy.

At the risk of stating the obvious, this article doesn't really talk about why they picked Clojure beyond one or two superficial reasons.

Ah, missed that.

Reminds me significantly of (and probably draws inspiration from) "The Gervais Principle"[1]

[1] https://www.ribbonfarm.com/2009/10/07/the-gervais-principle-...

Well, in the past we (a) died younger and (b) burned a lot of things for heat. I'd argue that literally nobody wants to go back to (a), and if every household in the world was still burnings a stove to stay warm we'd have an even worse global warming situation than we currently have.

This might work for you, but I'd say in general this sounds like a fairly unhealthy workplace. Poor engineering practices (when exactly do the tests get written if not day by day?) and negative co-workers? You have to distrust someone pretty strongly to call them a "hardcard manipulator"

I don't think of this as a hedge; rather, it's saying "look what I was able to do in my spare time!"

I think you're missing the point. It's not about MZ's motivations or even how FB thinks about things internally. It's about the perception by significant numbers of people - and, critically, their elected representatives - that FB is fundamentally shady.

There is a very good chance the path we're on ends with FB and potentially all of the social networks being considerably more regulated than they are currently, and I'd argue GDPR shows that we're already well on our way down that path.

Well, it helped that doing anything other than lying on my back triggered absolutely massive waves of pain. So I didn't really have a choice. I recommend taking a sleep aid beforehand and following all the usual advice for insomnia - avoid screens, read a book, lower the lights.

It felt very weird for about 3 nights, and after a few weeks started to feel natural.

I had to learn how to do this a number of years ago due to a severe spine injury. What I've found helps is having a large-ish but very soft pillow that accommodates my head lying to one side or another while I'm sleeping on my back.

I signed up for this a few years ago when I was working for a larger company. It was full of people using the anonymity to be very petty. I deleted my account without posting anything.

More details in profile - https://griffin.sh

Broken how? I got a bunch of signup pings off this thread - Bug report welcome founders@griffin.sh

Ah I hadn’t heard that. If that’s the case then I agree 100%

In fairness to him, the OCC (the national banking regulator in the US) has had a practical moratorium on new national bank licenses for most of the past 10 years. You can get a new state bank charter, but not a national one. I can't blame him for not wanting to start a bank in the US lately.

I can't tell if this comment is in jest or not, but we actually are starting a bank in the UK to be a platform bank for fintechs across Europe and beyond. Happy to chat.

Certainly possible now that the OCC has opened the door to a "fintech banking charter". In the past I'd have said this was impossible as Stripe has little interest in finding itself regulated as a bank or in becoming a bank holding company.

In Europe, PSD2 and the Open Banking initiatives provide this. All retail and commercial banks are required to provide open APIs to Account Information Service Providers or Payment Initiation Service Providers (the former being read-only, the latter read and write). Essentially you're talking about a delegated access / OAuth-style universe for banks and technology companies. It'd be great to see the US adopt something similar.

Fundamentally, banks are scale businesses (much like most technology companies). Expensive to start, but once you've got the product / regulatory infrastructure in place your marginal revenue is really low-cost.

We've actually started a new bank explicitly to provide the banking infrastructure to companies like this that are good at owning the UX advantage. Our conviction is that in the long term banks will be disintermediated from most of their customers and there will be a general unbundling of banking services - you won't get your mortgage / small business loan / checking account all at the same bank.

Also check out Griffin: https://griffin.sh

I think there's a bit of confusion around what market comp is. It doesn't matter that Facebook, Apple, etc. are willing to pay top of market at $500k - while they're big firms who employ a lot of people, I'd wager they're still not the majority of the labor market. It is also unrealistic to suggest that the "average" engineer has a high chance of being hired there.

I suspect true "market" rates are between $120-200k, depending on the seniority of the employee and the stage of the firm. A lot lower once you step away from US Pacific.

I think the best reason to be an early employee at a startup is if you plan on founding your own startup one day and want to get some hands-on experience. You won't make the strategic decisions, but you'll have easy access to the founders and will get to learn a lot about how to set up and run a company (both good and bad).

So why not just found your own company? Well, here are a few reasons: (a) you need financing but don't know how fundraising works and don't have a network to tap (b) you don't know how sales works (c) you don't have a good co-founder lined up (d) you're still a junior engineer and don't feel confident you could build a product by yourself (e) you haven't found a problem that excites you enough to feel justified working on.

I also notice a heavy trend in the comments here towards: employees should be rewarded almost as much as founders. I'm sorry, but no. A good founder will likely have spent years researching and understanding the problem space before finally trying to set up a company for it. They may be paying you with their own money. At the very least they had to starve long enough to raise enough money to pay you. They've taken on pain and risk. If you want to get equity comparable to a founder, go work for a pre-seed startup for free. In other words, if you want founder stakes, go be a founder.

As someone who has worked in an environment where thousands of machines were being deployed to, this is ridiculous. Orchestration is an area where good tooling makes an absolutely massive difference. Shell scripts are not a good tool for observability or handling rollbacks.

I really can't recommend picking two languages at once. Learning one new language often involves having to adapt to a significant number of new idioms that can really stress the mind - I've never been successful at diving deep into two languages at the same time. Far better to pick one language and really focus on it for 6-12 months. Then, at the end of that time, think about whether you want to work on it for another 6-12 months or whether you want to learn something else. Sometimes the language just won't take early on, and it's fine to pick something else then too.

He's referring to the time Simple started - in 2009, in the peak of the financial crisis and right after the Office of the Controller of the Currency initiated what is now nearly a 10-year moratorium on the granting of new national bank charters.

We came to many of the same conclusions.

This is why my co-founder and I decided to relocate to the UK in order to set up our bank. The friendlier regulatory ecosystem here is just a huge advantage compared to the US.

Plus, there's still a pretty clear path to getting a national US banking license - it just requires you to expand in from abroad, rather than trying to build it within the US.

For a seed stage company, 100k is not at all unreasonable. Extremely early employees get hefty stock grants to compensate for this.