randomgiy3142

Yeah my first response to 99% of these kind of interview questions is why? I look at ROI and how working code impacts budget and investment. It pains me to say it but the biggest companies out there have horrible code in COBOL that frankly doesn’t make sense to change it over. Similarly bug squashing usually is how did it pass QA/static analysis/dev ops.

You’re setting up a relay using two well known domain names it seems. And you’re encrypting files that probably can’t be decrypted using MITM so you’re sending all kinds of “red flags” if they use any number of MITM detection software.

To be fair our offshore team was so bad with security (“doesn’t work? Turn it off!”) it is unfortunately necessary. If I had a slightly different app “magick wormhole” they’re likely to use it if it had a pretty GUI.

Like if we didn’t have strict security policies in place how do you manage 500+ “developers” who have no repercussions? Part of it is getting the cheapest labor possible, part of it is security is hard to do right and part of it is english as a second language issue.

It is much easier to put everyone in an incredibly locked down environment than it is to have them decide what’s secure or not. If I were to fork this and internally use our own DNS and put a GUI wrapper and there’s a flaw in the implementation of magic wormhole I’d be in much more trouble than using Crowdstrike which no one will get fired for using for example.

Because translations are copyrighted so it is complex to get legal rights for them.

So have a good idea, be lucky and smart, and most importantly come from really rich families because floating 2.8million through two (!!?) crashes on an obvious now technology but log processing in the early 2000s is like describing how to win powerball.

Isn’t this what rpc/pbs/code-gen theoretically does? Markup/down sent? Then you don’t have to worry about the client or server language but settle on a subset of features somewhere.

My flow is that focusing on the data structure and methods allows the UI to be completely abstract. Not in a magic way but in a “maybe we should use cross compile this into C++ and use OpenGL if the UI becomes too complex. This is how you get AutoCad or film editing software running essentially as a Proton/video game runs.

When your app gets beyond a certain level of complexity (window composition, etc) yeah there’s ways to make it happen.

I tried explaining to people that you’re dealing with systems that are so antiquated places accept Diner’s Club cards. Accepting a credit card at all was a big deal because you literally copied a number and hoped it worked. People have cards that don’t have email associated with them. Furthermore there a ton of settling nuances. It’d be like building a browser if you were an alien who was given RFC specs.

I’ve worked with giant companies working directly with providers. Testing legalese and reality are far apart. In no scenario would we have the customer “test” a major new feature rollout. We’d have a budget and someone would make a real purchase then donate to charity the good or usually it was office candy for a month. I doubt the budget was even touched. We likely had provisions the prevented a $10k charge on a $15 product, that never happened. The only issue was that it’d skip normal QA (India has weird rules), and usually actually be a frivolous purchase or purchases on corporate and private cards.

Yeah do they check the tutorials are right? Seems like a horrible way of conveying information. We do similar things (in the open not documentation) and let analytics drive it. If a YouTube post gets nearly zero views but examples on github we change our priorities… awesome

Yeah but it turns so niche you write visual 3D interfaces in Unity for your specific application. Terminal editor improvements are visual programming.work iterm2 I can see command history and dig into problems.

There’s a huge advantage to see programming for non programmers. Do you know how hard it is to explain hypervisors, VMMs and OverlayFS to non programmers? I gave them a simple example they need to make a simple text update to a Photoshop file. It costs a renewed font license for the latest but essentially an application library they can open the legal version they bought on windows see it look sort of okay looking on osx make changes to the file and be done with it.

One thing ChatGPT taught me was people don’t understand code. SQL was supposed to be ChatGPT of the day. Executives had no concept the “power” of essentially an NLP on top of structured data.

And for developers a high level AutoCad like view of what’s talking to what, etc. is valuable. I don’t need to visualize a merge sort but maybe a realistic abstract demo non programmers and programmers can work against.

I got tired of corrupt paste programmers mocking up design in Figma. I want to show you acquired Product X on SAP and it’s a different version. Using token data I want to be like as of right now you see realistic data and users, zoom in, show it merge in a datalake and produce an auditable report.

And that it runs strategies so algorithms (probably AI) can show aggregated date and where things might be failing. Zoom out see two icons from two company’s on different software and a generic data lake, zoom in and see and update data. See a message in Skack/email/etc. indicating whatever. Show elementary that consultant leads products like Twilio run off AWS and telemetry shows they don’t use but one feature and we can swap it out. All ok a running system without awkwardly switching windows. It helps executives get the big picture,’itnactualu works and you drill down enough you see type forwarding and an IDL for the programmers.

When was the last time you used merge sort not in std library or visualization would have helped you solve it? And yes I’m not talking just about corporate software.

Just the other day I was wanting to show an overlayfs/squashfs. I mean I’ll go down to C or god help me ASM so it’s not technical ability. It’s usually not even big vs little indian that trips people up but the big picture.

I wish so bad things like Salesforce would release their code, it is very rare someone would “steal” it and companies I know would benefit greatly and still pay full cost if I could rip out things and see their documentation is wrong and not have to go to COBOL or disassemble to me that’s visual programming. “INT32 actually a mutable type oh yeah they also made up their own types. That’s in red. In fact here’s a point map of all types that don’t match specs. First question why (in 1994 a driver didn’t work right so we did some weird type forwarding) ok… but in huge systems I’m not sure visual programming would help. Maybe weird things like btree in rust is better until you make an roc call outside of rust for some reason?

Here’s what happened. I’m not in the VC world but work closely advising big companies, there’s a lot of cross over between a Honeywell and a VC. I’m a traditional hardcore engineer so I’ve seen it from systems level (kernel, fs) to web, AI, etc. so I’ve seen it all but luckily more as a consultant/architect who could you know code. Cloud computing became big but was expensive. It had obvious benefits but for ignorance or timing people went from in house server racks to same thing on clouds. This id important upgrades are expensive for mission critical software. A lot of initiatives aka budgets paid developers to implement things that weren’t related to do that. SaaS came along. Now you didn’t have upgrades but you still needed to adjust the software. That’s mainly config based and junior or offshore labor can do that. AI appeared and people thought this awesome. They didn’t realize people were smashing into got until something built. Also the mindsets of companies didn’t change. They only hired SWEs to maintain Oracle, not improve business. They saw what ChatGPT could do by asking it questions and thought oh I didn’t know computers could do that. When generating code has been done forever it’s not that’s hard it’s knowing what needs to be solved. So you have older execs thinking AI will need a bunch of cheap labor creating software not realizing we’ve already made software creation almost stupid easy with HLL. A lot of good ideas don’t come from SV stereotypes but there’s a lot to be said who open source or otherwise contribute now not having SWSlE jobs.

Anyway every company stole the NLP/LLM/Rag model out of in chat bots got better results no need for ui budget (looking at you salesforce). Some tasks like call centers were pretty much also made for LLM. Other AI advances like sorting packages and treading the line between algorithm and AI do well.

More importantly big companies are making tons of money and there’s not been appreciable advances in technology. LLMs will be local and integrated it won’t be called AI. Workflows will improve etc. right now it comes from big tech with cash on hand it looks like big iron from the 80s. The internet didn’t see anyone control it and was relatively cheap to get into. Create a 3d image scanner that can process millions of say walnuts on an assembly line as good or bad isn’t sexy and niche… and expensive.

So we are in a lull companies are bringing dev in house and aren’t being experimental. Unless you have one of 7 big tech firms behind you won’t get looked at. So right now it is get to the state in b2b to just get bought.

Also b2c is controlled underneath by AWS or a big cloud provider.

Good thing companies are realizing they don’t need Google levels of high reliability. I see a real progression back to on prem with slimmed down K8S variants with Chromebook like devices. Google spent a lot making those “just work” as anyone installing even Ubuntu can tell you. But we are basically at the point someone will create a business like Steak deck with hardware compliance. Servers where apps can be deployed like VDIs and a resurgence away back to guy in his garage software development. It’ll still need VC funding but with secure infra you can have on prem SaaS catering to the mid market.

We’d call these non reproducible non errors (200 for GraphQl) as “usefully wrong.” You see this in AI a lot where they spent millions in market research or ask MBA types how to recommend a product and it turns out that someone buying a new laptop tends to also want to buy new shoes. You’d get people like executives not happy they looked bad for spending money on one thing only to find out they were way off. Good news is more revenue makes people look good, so that’s not a hard problem. What’s hard is if a contact us form stops working (bad example), people stop using it but still use the app the same way and spend as much is the form and the people behind it necessary or are people brand loyal and willing to put up just with a minor bug? Similarly if we get something not working and can’t reproduce it did a network card have a low level error that propagated in such a way even our monitoring couldn’t pick it up?

I thought this was a new error in complex systems we have now with hundreds of clusters creating basically non-deterministic problems. But fine I remembered before kernels became better at talking to things like drivers and external hardware we’d see weird bugs outside our boundaries that were really hard to track down and often never manifested themselves in the same way. This is when you’d go to the weird guy no one talked to and in a week he’d have some piece of odd C code with a hex value doing logic no one understood that bypassed whatever error we were having.

It is too bad those guys that I’m pretty sure didn’t do much largely fell victim to the MBA thinking of the 90s. Now we usually will have one team go well we are calling the code right and the other saying they are sending it right and both aren’t wrong except it isn’t working so they are. We’ve reached a point where we have contracts with every vendor because the problem usually is actually like a Cloudflare :) but I’d argue it’d be far easier to just fix or create a work around and file a bug with them then spend more time on daily calls working with someone like you and knowing your progress. So I know what you mean by tools companies use. Unless it has hit industry standard we won’t even evaluate open source as we couldn’t blame someone.

Nah Linux still has problems that Windows had years ago: supporting everything with a mess of drivers. It works well if done in an Apple model of software and hardware all integrated. Chromebooks get close to this.

Don’t get me wrong I use Linux every day, it has come a long way. One bad package update can kill an rpm-stree setup (not mentioned) and the boot isn’t partitions if you use ZFS these partitions aren’t needed.

If you’re okay with an iPad like locked down setup it works really well. But why not just buy an iPad? I mean it doesn’t look like it did 10 years ago but the OS is free basically.

I’m very pro-Linux but from a hardware maker point of view it is minuscule so it is still in power user territory.

I use zfsbootmenu with hrmph (https://github.com/leahneukirchen/hrmpf). You can see the list of packages here (https://github.com/leahneukirchen/hrmpf/blob/master/hrmpf.pa...). I usually build images based off this so they’re all there, otherwise you’ll need to ssh into zfsbootmenu and load the 2 gb separate distro. This is for home server, though if I had a startup I’d probably setup a “cloud setup” and throw a bunch of servers somewhere. A lot of times for internal projects and even non-production client research having your own cluster is a lot cheaper and easier then paying for a cloud provider. It also gets around when you can’t run k8s and need bare metal. I’d advised some clients on this setup with contingencies in case of catastrophic failure and more importantly test those contingencies but this is more so you don’t have developers doing nothing not to prevent overnight outages. A lot cheaper than cloud solutions for non critical projects and while larger companies will look at the numbers closely if something happened and devs can’t work for an hour the advantage of a startup is devs will find a way to be productive locally or simply have them take the afternoon off (neither has happened).

I imagine these problems described happen on big iron type hardware clusters that are extremely expensive and spare capacity isn’t possible. I might be wrong but especially with (sigh) AI setups with extremely expensive $30k GPUs and crazy bandwidth between planes you buy from IBM for crazy prices (hardware vendor on the line so quickly was a hint) you’re way past the commodity server cloud model. I have no idea what could go wrong with such equipment where nearly ever piece of hardware is close to custom built but I’m glad I don’t have to deal with that. The debugging on those things work hardware only a few huge pharma or research companies use has to come down to really strange things.

There are bigger things than a failed trial. There was a lot of money invested in this and when it was found to not be harmful there was a large push to show success. A lot of careers will be cut short because of this. There’s a lot of motivation to show something works. I was told to avoid research because of this kind of thing. There were a lot of highly paid people in a room that were not scientists but tasked with showing how much money could be made. That sounds evil but their intentions are good, that doesn’t help ALS people but there’s enough money put into this I wouldn’t be surprised if some progress is shown somewhere.

A good interview should be a good fit for both parties. Just have a casual lunch with them. You should be able to tease out if they are a good fit. These “hacks” are so ridiculous. Way to overthink something.

I am not Indian but I work for a large Tata like IT firm. This hit way too close to home. There a lot of cultural issues here that comes down to management being rewarded if things are done cheaply and discouraging any agency or self-realization by the developers. If I saw this in the US, I’d walk out. They literally don’t have that option as there’s a 90 day salary clawback if they do. Some general thoughts:

- Most management has a non-tech background. So they get what they want to hear and don’t want to hear what’s wrong.

- Thinking this coming from the same team or from the same company is wrong. They silo developers like crazy. There likely was an API developer, an Office 365 developer, frontend developer (so specific it is down to the framework or stack!) and the developers themselves will not touch anything they aren’t “certified “ in.

- I have been in meetings on $100 million projects where they will seriously argue over the cost of sendgrid. Eventually this will come down to no one having “sendgrid experience” And some developer saying they can do it in Office365.

- Security team will get the first cut in budget since it should “already be secure.”

- You are likely talking over the head of the nephew hired to do security for this. Will the government or anyone sue them? No, so why is this guy bugging us.

Developers aren’t encouraged to develop but get tickets out and not question them. The manga is “It wasn’t in the requirements” all the way down the chain.

I work with smart developers out of India but it is not a culture of innovation. This kind of work is treated like a call center. Don’t go off script, stick your little problem domain, if we aren’t failing we are winning.

Once you get into a mindset of quality not mattering and only pushing down the costs there’s no coming back. Look a large retail brands and how disposable we treat things. They went from northeast to southern US to Mexico to China. The creation of the product will happen in the US for cultural reasons (you cannot outsource marketing generally), but it all hits a wall.

You’re missing a huge point: there’s a ton of devs that went into react bootcamp or whatever. Building a 50 person team on a technology that isn’t widely used or heavily supported by Google/Adobe/Microsoft is a huge hurdle. Even Typescript isn’t really used correctly, people basically use ts-lint on a lot of projects I see. For large companies they will also turn projects into BAU (business as usual) where the pressure is to push down costs and move it overseas. Finding a large overseas team on htmx at React prices will be difficult.

Typescript is popular because people don’t use it a lot, and it doesn’t break things.

If you’re introducing something new the bar is incredibly high to be obvious as to why to switch.

While true this is divorce from organization’s hiring practices. I have to fill out a form to request someone and htmx won’t be on it. You will get someone who only know react and will stomp their feet as to why can’t we use react. A lot of bootcamp programmers are cert jockeys who don’t care or know about development, not that they can’t learn.

What else are you going to do with them? Subsidize businesses through tax breaks or pay to upkeep them through taxes are the only answer. It is incredibly expensive to turn office space into residential space. Ideally any new office space or residential would require a plan to convert to one or the other before being approved, but who saw WFH being so widespread 30 years ago?

The worst scenario is that businesses en masse begin breaking their leases and the buildings quickly fall into disrepair. I’m sure there’s people working on a solution to move all the plumbing and infrastructure out of a central core to make better use of these structures. It might be best to keep things as they are until that happens.

I’d be careful with Hetzner. I was doing nothing malicious and signed up. I had to submit a passport which was valid US. It got my account cancelled. I asked why and they said they couldn’t say for security reasons. They seem like an awesome service, I don’t want knock them I just simply asked if I could resubmit or something the mediate and they said no. I don’t blame them just be careful. I’m guessing my passport and face might have trigged some validation issues? I dunno.