I wrote the below to explain to our users what was happening, so apologies if the language is too simple for a HN reader.
- 0630, we switched our DNS to proxy through CF, starting the collection of data, and implemented basic bot protections
- Unfortunately whatever anti-bot magic they have isn't quite having the effect, even after two hours.
- 0830, I sign in and take a look at the analytics. It seems like <SITE NAME> is very popular in Vietnam, Brazil, and Indonesia.
- 0845, I make it so users from those countries have to pass a CF "challenge". This is similar to a CAPTCHA, but CF try to make it so there's no "choosing all the cars in an image" if they can help it.
- So far 0% of our Asian audience have passed a challenge.
Yup. AWS went down at a previous job and everyone basically took the day off and the company collectively chuckled. Cloudflare is interesting because most execs don’t know about it so I’d imagine they’d be less forgiving. “So what does cloudflare do for us exactly? Don’t we already have aws?”
This ad nonsense aside, don't buy Samsung refrigerators. They are so awfully made and difficult to service that almost no appliance repair companies will touch them. I got suckered into buying one a few years back and it was awful. The ice maker didn't work, every few weeks I would sop up a gallon of condensation at the bottom of the cheese drawer, and eventually it just died. I went to a local appliance store and they chuckled when I told them. They would never carry that brand. Just fridges, don't want to talk about other appliances.
This was downvoted? That’s sad. I tell folks getting started that they should build something without any framework so they can feel the pain that things like React are built to solve. Understanding the fundamentals is important so you can look at something like hooks as cognitive overhead that needs to be accepted
Ex employee here who built an aws service. Dynamo is basically mandated. You need like VP approval to use a relational database because of some scaling stuff they ran into historically. That sucks because we really needed a relational database and had to bend over backwards to use dynamo and all the nonsense associated with not having sql. It was super low traffic too
Agree on the importance of testing. Among the production-grade codebases I've worked on, I've found that the dynamically-typed ones have more comprehensive testing. It's just so easy to think that a successful compile means your code will work. I've also found that it's harder to set up great test systems for static languages because you often have to modify your logic just to make it testable (looking at you, IoC). A delightful test system is one that engineers will use during development because it saves time, not an afterthought. For whatever reason, I haven't ever found something that provides this type of experience out of the box. In one organization, we spent months building a rig and achieved true TDD. The result was fewer production issues, faster development, and of course, better test coverage numbers. We eventually switched from javascript to typescript, but it didn't compare to the difference that test system made.
I'm working on a system that helps surgeons make precise bone cuts during knee replacement surgery. Believe it or not, manual cuts are still the standard in that type of procedure. Robotic systems exist but they are very costly, big, and actually add time to the surgery (bad news when you are under anesthesia and your leg is in a tourniquet).
It uses 4k stereoscopic capture and bunch of ML models to match bone position with sub-millimeter precision. The surgeon screws a metal base piece into the bone, and we detect where that is in space. Then, a Stewart Platform adjusts another part that is placed onto the base. The robotic adjustment allows the base to be placed in a ballpark area, with the robotically-adjusted piece oriented in the exact spot where the surgeon needs to cut.
The net result is a robotic system that is many times cheaper than the least expensive incumbent, decreases surgery time significantly, reduces error, and basically "just works" as opposed to requiring a ton of training. We are debuting at a tradeshow in October.
This sounds awesome! Can you tell me more about what kind of expertise do you need to develop such a system? As in the most important knowledge one most have to be able to work on such a thing
Thanks, buddy! I'm having fun. There are a few sides to it. There are the actual physical surgical tools that you have to design, test, and manufacture. Then there's the robot that adjusts those tools. That stuff is a lot of CAD and 3D printing. The camera is a big deal and it's a ton of work to get that right. Then of course you have all the software, which is a slew of computer vision models that operate on a local computer in a careful dance of resource orchestration. The software has a lot: UI, grpc services, ml models on containers, inverse kinematics for calculating robot position, hardware interfaces, etc. Then there's a bunch of regulatory, validation, compliance, etc.
To answer your question about expertise, it really depends on what you are interested in. We have some dedicated mechanical engineers with medical device experience. The software is handled by a few computer vision and full stack folks. So there's different skillsets.
I'm a bit of a journeyman and as a result, I am decent across all of it. I always did software and went where the wind blew. It's been 20-something years since I graduated so I've seen a lot. About 10 years ago I got a job I was totally unqualified for, which was R&D for a company that made lab equipment for testing gas and oil. I was solo and had to learn all the mechatronics stuff - CAD, microcontrollers, electronics, etc. Check out this video: https://youtu.be/MA6hnyXx4p4. That specific experience allows me to be the glue in our engineering org.
To work here, you don't need medical experience. We have plenty of that. One of the cool things about engineering, especially software engineering, is that you can float around between verticals. I've learned all about media, finance, petroleum, insurance, waste disposal, etc. The skills translate. If you are purely software, I recommend picking up an Arduino and some motors and building something like a simple pan/tilt mechanism with an accompanying mobile app. Just do it. It might inspire you. I think curiosity and enthusiasm are the most valuable traits one can have.
That seems like a long journey. I think playing with Arduino is a perfect way to get started on the interaction between software-hardware. Thanks a lot for your answer!
VR training hasn't really stuck in ortho training. It's legit and people do it, but it's a distant second to doing it live. I haven't seen that much of it, but from what I have seen, the fidelity is quite low. It would be cool to see what a good game studio could produce.
Doctors, like most people, don't like stuff on their head. Plus in ortho there is a lot of feel to it. It's often referred to as "carpentry". The docs I know, especially those with experience, would prefer a video and a cadaver lab. Even that's a lot to ask because they are so swamped. In every surgery there is a rep from the implant company, and those reps are really the ones doing the training.
So there is certainly potential but it's just not to the point where people are excited about it.
Used to work for a major implant company, we had a cadaver lab at company headquarters and flew in doctors for training. The rep in the OR is mostly just for support. They carry in a bunch of hardware in case the doctor needs to pivot midway through, they may need different sized screws or something. They can show doctors how the device is meant to connect together without bone/tissue but they often have little to no medical training, they are sales representatives.
Hey there, thanks for writing. For those who don't know, the sales reps have to carry in racks and racks of parts. Most go unused but the last thing you want is something to go wrong and not have the piece you need readily available.
I've been in many different ORs and I've found that the rep's knowledge and level of participation varies a lot. Some of these reps have been working with the same products for years, the doc fully trusts them, and could probably perform the procedure themselves. Others not so much.
This sounds about right. My time in the industry, I always remember our top selling rep was basically royalty at the company he drove so much volume - but, he was only about 22 and never finished college or had any actual medical training. He just grew up with a surgeon Dad, country club, etc. and had deep connections with a lot of surgeons by virtue of this. He was also very charismatic and a good salesman, just not the guy you'd want training your surgeon before they fused your spine.
I understand your point and if I were the author I would want either a disclaimer or a fix. File an issue or make a pr. Filing an issue is quicker and more fruitful than dealing with folks here
As a startup you wanted to underpay people that you import from overseas? There's a lot about Trump immigration policy that I do not like but this one I think is totally appropriate.
If you read the articles you'll see that even Microsoft is telling people to get back to the US to avoid this fee. Not even big companies want to pay it. You have forgotten what the H1B visa was supposed to be for. It was to bring in experts that could not be sourced to locally. What it's turned into is people paying for cheap labor from overseas to come here. There is lots of locals with lots of experience and knowledge and skills but too many companies are unwilling to pay the wages that they're worth. So they'll bring in subpar people, who are not experts in their field, to do the work. It leads to a mediocre work product as well.
Because honestly, if they were really an expert with domain-specific knowledge that is difficult to source locally an extra hundred thousand dollars is pennies compared to the profit that they would bring.
No, I don't want to underpay imported people. I'm not talking about teams of low-skilled people. You don't know my situation, so I'll explain. I work in a startup that requires a specific ML computer vision skillset. And we are in the surgery space. I found a brilliant person who is a master at computer vision, loves the medical field, and is willing to take a +6 figure salary and equity. I tried to find someone like that stateside and couldn't. He wants to live here and would be a great addition to the country. He's in Canada now, has been for years. Unfortunately he's still there now because of the recent travel ban for those born in certain countries.
Thing is, I would pay $100k extra to have him in office. In my case it's moot because of the travel ban. My point was, this whole thing simply benefits the companies that can afford it - google, meta, amazon etc, who happen to be big contributors to the administration. $100k will not stop them from hiring foreign talent, but it may stop companies like mine. Flagship companies might be psyched about this.
I know exactly what your situation is. You're looking for H1B Visa workers which are intended to be highly specialized technical skills that are not locally available in the United States. If a $100,000 fee to bring in a highly technical specialized worker that cannot be found in the United States is too much then you are not looking for a highly specialized technical worker with domain specific knowledge that doesn't exist within the United States.
It's really as simple as that. The H-1B visa system has been abused repeatedly by many corporations and you yourself are seeking to abuse it. I have seen many H-1B visa holders and none of them seem to have highly specialized domain-specific knowledge. I'm sure there's a few that do but the vast majority are average computer science graduates from India that get imported. They do average work and there is nothing special about their knowledge or their skills that they're bringing to this country.
So hire people here first. You absolutely are going to need to pay a US citizen a good 30 to 50% more than what you're going to pay that mediocre H-1B person because if you were looking for a highly specialized person with domain-specific knowledge then you would pay the US person or the H1B person the exact same wage. This fee for H1B Visas is now going to make these companies, and yourself, look very seriously about your hiring practices and your abuse of the system. Because if you need the specialized person the cost is insignificant. Because a one-time $100,000 fee is pennies compared to the salary that you should be paying yearly which is $200,000, and that 200k is not including all of the cost of employment and taxations that you have to pay.
So there's massive holes in your entire story. You're either not seeking skilled domain specific knowledge people or you are absolutely underpaying people who are highly skilled with domain-specific knowledge.
You're being too aggressive for my taste and I don't think there's any dialog to be had here. I was wrong though, this particular visa was O-1 and not H-1B and they are quite different. I was discussing a single person who was uniquely qualified. And passed the rigorous O-1 application process only to be prohibited from crossing the border because he was born in a particular place.
I'm a startup with modest angel funding. We can't compete with larger companies salary-wise and so far I haven't been able to find competent people who are willing the play the startup game when they can get half a mil straight out of school. We have only a handful of software engineers. But I don't have to tell you because like you said, you know exactly what my situation is.
This clearly isn't a productive conversation so let's just drop it. Enjoy your weekend!
So now you're just changing the subject. Because we were talking about H1B visas and now you're complaining about someone with a different visa type being prohibited from crossing the border. Maybe what's your sensing as aggressiveness is me just pointing out that you don't seem to be talking in good faith or you constantly keep changing what you're talking about without indicating you're actually changing it. Makes it quite confusing to have any kind of cohesive discussion.
I think if you'll refer back to my original response I acknowledge that I do not like The vast majority of immigration related policies that have been instituted under the Trump administration. It was only this one specifically about H1B visas that I do agree with. So you were apparently talking about not H1B visas which resulted in this confusion.
