git workspaces, combined with some guard rails to keep the llm on task. For example a detailed document with requirements and implementation details and a unit tests that the llm can keep verifying against and/or enhancing between each step to increase the reliability of it’s output.
Two days after they released their web interface for Claude Code I was hooked. Haven't really used the regular interface or the app since. Oh god, I'll never go back to copy-pasting code.
This might be off topic since we are in topic of AI tool and on HackerNews.
I've been pondering a long time how does one build a startup company in domain they are not familiar with but ... Just have this urge to 'crave a pie' in this space. For the longest time, I had this dream of starting or building a 'AI Legal Tech Company' -- big issue is, I don't work in legal space at all. I did some cold reach on lawfirm related forums which did not take any traction.
I later searched around and came across the term, 'case management software'. From what I know, this is what Cilo fundamentally is and make millions if not billion.
This was close to two years or 1.5 years ago and since then, I stopped thinking about it because of this understanding or belief I have, "how can I do a startup in legal when I don't work in this domain" But when I look around, I have seen people who start companies in totally unrelated industry. From starting a 'dental tech's company to, if I'm not mistaken, the founder of hugging face doesn't seem to have PHD in AI/ML and yet founded HuggingFace.
Given all said, how does one start a company in unrelated domain? Say I want to start another case management system or attempt to clone FileVine, do I first read up what case management software is or do I cold reach to potential lawfirm who would partner up to built a SAAS from scratch? Other school of thought goes like, "find customer before you have a product to validate what you want to build", how does this realistically work?
I think if you have no domain expertise or unique insight it will be quite hard to find a real pain point to solve, deliver a winning solution, and have the ability to sell it.
Not impossible, but very hard. And starting a company is hard enough as it is.
So 9/10 times the answer will be to partner with someone who understands the space and pain point, preferably one who has lived it, or find an easier problem to solve.
1. Compliancy with relevant standards. HIPAA, GDPR, ISO, military, legal, etc. Realistically you're going to outsource this or hire someone who knows how to build it, and then you're going to pay an agency to confirm that you're compliant. You also need to consider whether the incumbent solution is a trust-based solution, like the old "nobody gets fired for buying Intel".
2. Domain expertise is always easier if you have a domain expert. Big companies also outsource market research. They'll go to a firm like GLG, pay for some expert's time or commission a survey.
It seems like table stakes to do some basic research on your own to see what software (or solutions) exist and why everyone uses them, and why competitors failed. That should cost you nothing but time, and maybe expense if you buy some software. In a lot of fields even browsing some forums or Reddit is enough. The difference is if you have a working product that's generic enough to be useful to other domains, but you're not sure. Then you might be able to arrange some sort of quid pro quo like a trial where the partner gets to keep some output/analysis, and you get some real-world testing and feedback.
I think it comes down to, having some insight about the customer need and how you would solve it. Having prior experience in the same domain is helpful but is neither a guarantee nor a blocker, towards having a customer insight (lots of people might work in a domain but have no idea how to improve it; alternatively an outsider might see something that the "domain experts" have been overlooking).
I just randomly happened to read about the story of, some surgeons asking a Formula 1 team to help improve its surgical processes, with spectacular results in the long term... The F1 team had zero medical background, but they assessed the surgical processes and found huge issues with communication and lack of clarity, people reaching over each other to get to tools, or too many people jumping to fix something like a hose coming loose (when you just need 1 person to do that 1 thing). F1 teams were very good at designing hyper efficient and reliable processes to get complex pit stops done extremely quickly, and the surgeons benefitted a lot from those process engineering insights, even though it had nothing specifically to do with medical/surgical domain knowledge.
Anyways, back to your main question -- I find that it helps to start small... Are you someone who is good at using analogies to explain concepts in one domain, to a layperson outside that domain? Or even better, to use analogies that would help a domain expert from domain A, to instantly recognize an analogous situation or opportunity in domain B (of which they are not an expert)? I personally have found a lot of benefit, from both being naturally curious about learning/teaching through analogies, finding the act of making analogies to be a fun hobby just because, and also honing it professionally to help me be useful in cross-domain contexts. I think you don't need to blow this up in your head as some big grand mystery with some big secret cheat code to unlock how to be a founder in a domain you're not familiar with -- I think you can start very small, and just practice making analogies with your friends or peers, see if you can find fun ways of explaining things across domains with them (either you explain to them with an analogy, or they explain something to you and you try to analogize it from your POV).
Calling it a payment processor is a bit of an issue. The term "payment processor" has a very specific meaning in the world of online payments, and this isn't it.
This is a frontend for Stripe, so it's even a bit of a stretch to call it a Payment Service Provider. Payment gateway maybe, but really it's a Stripe frontend.
off topic: My goodness, anyone feel the UI/UX of the website to be really refreshing? I've lately been digging compact/industrialized looking UI vs 'touch'/comfort view that take up way too much real estate.
In the .NET ecosystem, I have noticed people to shame .NET MAUI because Microsoft themselves don't use this framework - Microsoft Team is built on Electron and not MAUI.
Why build a product on MAUI when Microsoft aren't too sure about it.
The answer to that is well known: Windows division builds WinUI/Win32 as their native C++/COM API, Office division went to React on their path to the web and the dev division fills gaps (WPF) and provides tools for external and internal devs (Maui for cross platform uis).
It is history not the lack of will. At one point the windows division was in shambles (remember vista) and WPF pops up. At another point, the windows and dev division have no answers to the office group (because you know who uses non win tech) so they went react. And then external devs screamed: where is the .net cross platform story so Microsoft acquired xamarin and later form Maui out of it.
It is history not lack of trust. But the outcome is the same: lackluster support for all UI toolkits.
Well summarized, and just as shocking today as it was every minute while it developed.
Someone needs to remind those cats that they own the platform. Being able to sanely develop apps for and on that platform should be possible, and UI kinda-sorta matters for that. At a certain point with the MFC they had it dialled in, while pioneering asynchronous browser tech, with many best in class tools. Decades later with a cross-platform cloud-centric stack they have a shrug emoji as big and wide as the eyes can see, and no sense this basic question of development will ever get improved.
Ballmer chanting ‘developers, developers, developers …’ springs to mind.
Ballmer was not my favourite person to run MS, but that was a pretty good time to be a Windows developer if you bought in. Early Nadella days too. Now? It's too easy to find no MS answers for almost everything, even triple-A gaming is not the windows anchor it's been since they ran that doom demo
> but that was a pretty good time to be a Windows developer if you bought in
I did, it was great. Very apple-esque in a way. As long as you stayed in Microsoft's garden, you had a good time. Microsoft had, at the time, one of if not the most productive stack to build GUI desktop line of business apps. If your whole org was Microsoft from top to bottom, even better. AD auth in your desktop app in a couple of lines.
It was an expensive stack for sure, but I'd argue there's still nothing that has come close to it if you want to build an enterprise desktop app.
As I understand it, Sinofsky is largely responsible for this mentality in the various places he worked across Microsoft: he instilled a distrust for anything originating out of other teams.
These days, I think Microsoft's web-based desktop apps mostly use WebView2 directly instead of Electron, so they don't have to bundle a browser. I think for Teams it happened at the same time that they moved from Angular to React.
The point about them not using MAUI still stands though. From what I understand, the .NET world has either adopted different abstractions like Avalonia, or stuck with tried and tested solutions like WinForms with proprietary controls. After all, they've seen this before with WPF which was never fully adopted by MS either, or with the debacle around Metro/WinRT. You're never quite sure what Microsoft wants you to use or will support in the long term. They also make Blazor, which is a different (and likely more accessible) way to build web apps with .NET.
Since we're on the subject of companies not dogfooding their shiny tech, is Google really using Flutter for their own apps? I feel like the evolution of the Android ecosystem towards Kotlin and Jetpack Compose implies otherwise.
Ive been building these apps (cross platform web based ui, C# backend) for years, and yes its finally good to see MS catch up and validate the architecture ive pushed since Xamarin. I wrote once wrote an electron version of this archand thought wtf are people doing? Things can be so much easier when you use a platform that knows how to multi-thread. At stages i had to build adapters/upstream patches for Chromeiunium directly onto Mac and Linux, and its was a major pain having to debug C calls.
Ive been using the same framework now for 10+yrs on apps in the stores, i wrote a small layer infront of the webviews and can swap out webkit, chrome, edge on demand. You really dont need much, just a constand way to boostrap logic and UI. 90% of code is shared across all platforms, there are def differences in WebView engines that you sometimes come across but those parts just get swapped out with browser specific JS. Ive found bugs and worked with browser teams at all vendors doing this and to see how simple this is with Dotnet these days compared to when i started is refreshing. Its easily the most stable cross-platform framework around, if you are stuck using something like flutter i pitty you, its just eletron with another skin. I can swap out and integrate directly with OS libs when i need to do stuff that the dotnet team hasnt gotten around to yet without re-writing. This has mean i really havnt used MAUI at all, but if i need to or could take advantage of it i can mix it on an Ui element by element basis. I prefer webUIs though, i have the chose to handle anything with either JS, WASM, or a combination. I can use traditional JS frameworks or traditional Native UI frameworks.
If i had started this process later, avalonia seemed to have the closest thing i required. It was just a bit a more complex /based on WinUI (which i dont really enjoy) but it supported all platforms and gave lower level api access. MS were smart, that canabalanised all these community effort and brought them into the fold. Every dotnet webview impl was a successfull community driven project before. They didnt write anything themselves from scratch.
> Microsoft Team is built on Electron and not MAUI
Microsoft Teams was released in March 2017. .NET MAUI was released in May 2022. In 2021, Microsoft replaced Angular with React and moved away from Electron to WebView2 (using the OS' built-in renderer rather than a bundled Chromium). So even the rewrite was a year before MAUI (and they probably started the rewrite before 2021). Plus, part of the point of using React there was that they could basically replace Angular bit by bit.
Microsoft Teams is just older than MAUI. It's like asking why Hadoop is written in Java and not Go or Rust or why Kafka is written in Scala and not Kotlin. Kafka was open sourced in January 2011 and Kotlin came out in July 2011. Kotlin wasn't an option given that they were developing it years before the language was released.
That's not to say that Microsoft's attitude toward MAUI doesn't leave concerns. There was some news a while back about a bunch of layoffs around MAUI. It's always concerning when there doesn't appear to be any dog-fooding going on - is this just some junk they're throwing at us that they don't want to use? I think some hesitation also comes from the Blazor side where it's looking like Microsoft doesn't really see Blazor as a React competitor so much as a way for internal company apps to be made quickly - in contrast to the Google IO presentations on WASM support for Dart/Flutter where they were emphasizing better-than-JS performance.
That said, Microsoft hasn't really released a lot of new (green field) stuff over the past 2-3 years. What product should they have made in MAUI, but didn't? You can't say Teams because that was a giant product way before MAUI even existed. Most of what Microsoft is doing is work on existing products - things they released before 2023/2024 and were in development before MAUI existed. Flutter had a 5 year head start on MAUI.
But there certainly is a feeling that Microsoft doesn't feel committed to it or at least not enough to put its weight behind it.
Never build a frontend on a .NET technology. Period. They always end up unsupported in the end. Just use standard web technologies and thank yourself later. I've been a .NET dev for a decade now and that's what I've learnt.
Blazor is pretty great. It is mature at this point and MS is using it internally more and more. Trying to go back to something like React makes me shudder. It's not perfect, but it's better than many alternatives.
I agree. For Blazor there is hope. It is standard based (web assembly, HTML, css) and it feels very intuitive particularly when compared to other spa frameworks like react. Also you can reuse all your html, css and design systems you have. Which is huge because like that it hooks up with the whole web development stack.
Tauri is pretty awesome. Rust backend, WebView front end. Nothing uses native desktop elements of course.
To be fair, there is no practical way to write native desktop applications using stylistically consistent UI elements AND have it be portable AND in a language that you enjoy using.
As far as I can tell, Windows 11 doesn't even have a toolkit with platform UI elements.
GTK on Gnome is pretty okay and GTK-rs is not dissimilar to React. Who know what MacOS uses but something something Swift XCode.
But I agree, just use web technologies. Write once, ship everywhere (and hum loudly when people complain about poor performance - joking, it's the vendors' fault we have to use web technologies).
In my experience with Tauri, it's pretty good on Windows, but not so much on other platforms, especially Linux. The decision to target different browser engines on each operating system means you still have to deal with a bunch of different OS-specific bugs.
For Windows you're dealing with Edge (so Chromium), on macOS you have Safari, and on Linux you have WebKitGTK. WebKitGTK has honestly abysmal performance, and you're missing a lot of modern standards.
The Tauri devs are looking at bundling a Chromium browser with it to deal with that, but that's still some time off, and leads to the same issue Electron has, where you have large bloated applications.
I don't know much about it but it seems like a weird combination. If you want high performance and low memory usage, you don't want HTML, if you want fast code writing, you don't want Rust.
I don't believe you generally end up writing a lot of Rust with something like Tauri. It is mostly web dev. While it is true that browser based UIs are slower than native, it isn't clear that .NET based UIs would be any faster while being very niche.
I could see a case where the core logic needs to be performant, but the UI does not. The front end could be some menus, displaying (not a giant amount of) data, and a progress bar, while the back end does heavy computing.
And furthermore, if you want fast code writing, you write in the language you already know. For some people, that is Rust.
- HTML is the main way of designing interfaces (whether we like it or not)
- Rust is the main language promoted by intelligence agencies with multi-billion dollar budgets because it is laced with their backdoors (whether we like it or not)
As wrong as it feels to have to use Electron for a desktop app, it really is the safest approach for most applications.
Qt also seems to be a good option, though there are licensing considerations for commercial applications.
I’m excited for various upcoming Rust options as well, but right now Electron is the battle tested option.
I am curious though about Avalonia. I’ve heard good things, but it’s definitely a smaller player compared to Electron. I’d most likely choose it over Microsoft’s first party frameworks.
Does their market share back up your take of them as horrible apps?
Are there QT or GTK competitors crushing them?
I always hear how terrible electron apps are, but the companies picking electron seem to get traction QT or other apps don't and seem to have a good cross platform story as well.
Users will happily deal with a suboptimal experience as long as there are other things attracting them to the product. That's why Microsoft can do whatever it wants with Windows without worrying their users will run off somewhere else. So if you care more about people than businesses, maybe it shouldn't be an excuse to pick "better dev experience" over the user's.
They said horrible user experience apps, not horrible apps. You can still deliver an app with a horrible user experience and build a profitable business. Ever done an expense report?
Companies aren't picking Electron due to inherent shortcomings in other platforms, they're picking it because it's easier (and cheaper) to find JavaScript devs who can get up to speed with it quickly.
Discord, VS Code, and Figma are all apps that individuals choose and are well liked despite many alternatives. Slack too I think, though I don’t have experience with it.
Your comment applies to Teams and I’m sure other electron apps. But the sweeping generalization that electron apps have terrible user experiences is pretty obviously incorrect.
Oh yes, the great old "works for me". On a yesterday's supercomputer, I presume? I live in a "developing" (have doubts it's really developing) country, most people are running laptops with no more than 8 GiB of RAM (sometimes it's 4 or less), and all this Electron nonsense is running like molasses, especially if you're trying to use a computer like a proper computer and do multitasking.
And most of the world is like that, very few of us (speaking globally) have $2k to drop on a new supercomputer every few years to run our chat applications.
I think Qt really is 'just' missing more language bindings, and a better hot reload story for more people to use it. Lots of devs (specially Free Software devs) would prefer to use native toolkits, if the prototyping experience was similar to how Vite is for web frontend stuff, I think Qt would be used a lot more.
Thanks for sharing. I hope that initiative goes well. And again, integration with whatever solution the language has for hot reloading is a must these days for UI development.
Same for me. I used to do desktop dev with MFC and WPF but these days you would be suicidal to build any app that needs to last for a few years on .NET. If totally needed, WPF is still the best bet. Otherwise I am pushing everything I can to the web.
You are right that WinForms and MVC have been around forever. However, Microsoft has continuously told devs that they are the past. So, you would be forgiven for expecting them to go away.
WinUI is the current official desktop paradigm and it is basically UWP from an API point of view. So the idea that UWP went away is not 100% accurate either.
Microsoft does not really abandon their UI tech like people say they do. But look how many different frameworks they have.
All of the above is Windows desktop only. There are a completely different set of UI technologies for the web.
UNO Platform (Open Source) allows you to use the WinUI API to target almost anything.
.NET MAUI is the official "cross platform" UI tech from Microsoft. It is what you use to target iOS and Android. As a bonus, you can target macOS and Windows too. On Windows, it uses WinUI. You will notice that the Linux desktop is missing from that list.
Here comes Avalonia to build MAUI on top of the Avalonia framework. This adds Linux and WASM to the list of platforms that MAUI will run on. Adding Linux is awesome. A lot of people have wanted that and it really completes the MAUI cross-platform story.
Adding WASM is neat but MAUI was never meant to target the web. If you use it for that, it is literally just the modern version of Silverlight. But Microsoft did not design it for that at all. It is just a back-end that Avalonia supports.
MVC is a design pattern, ASP.NET MVC is a framework that used MVC as its go to pattern. But MVC is not in any way only ASP.NET MVC. There are plenty of other UI frameworks that use MVC and the Wikipedia article lists a lot of them for example: https://en.wikipedia.org/wiki/Model%E2%80%93view%E2%80%93con...
No kidding - kind of wild that winforms is still kind of a gold standard experience today! I actually liked VB Forms - lots of easy rapid application development was possible.
Delphi was the best RAD tool though. It was native code, not a weird interpreted or jitted app. It could also build to a single exe file. VB struggled with an unwieldy engine for most of its life.
I like WPF and I code with it regularly, but the drag and drop UI builder was the worst aspect of WPF and generated terrible Xaml that was almost impossible to maintain.
I was going to suggest the same, just use WinForms. It's basically feature complete, and it's going to be the last UI framework Microsoft is going to yank out from beneath you.
Generally speaking, I wouldn’t take what Microsoft uses as guidance nowadays, given a lot of the software they produce. (This is not an endorsement of MAUI.)
That has been a problem since forever. Microsoft themselves rarely used the tools they gave to developers. SourceSafe, MFC, WPF and the .NET frameworks that followed were only for 3rd party devs. And when they used these tools, the software usually got worse. One example was Visual Studio. 2008 was really nice with great customization and good performance. Then they wrote 2010 with MFC and it was slow and lost tons of features.
I think it’s better on the server side with ASP.NET.
As far as I have heard MAUI is pretty buggy and has lost momentum. It will probably go on the long list of basically abandoned .NET UI frameworks
You are mixing your UI frameworks and versions. VS 2010 is written in WPF. WPF is / was Windows Vista's and 7's UX. Old Control Panel in Win 10/11 still is WPF. All the wizards like ClearType wizard is WPF. MFC is much older (1992).
Unfortunately Microsoft likes to jump into bandwagons and many engineers at the company seem to like to reinvent stuff rather than adopt. WPF, WinUI2 and WinUI3 all share the same Xaml based structure. So they could have adopted WPF.
It is not that Microsoft doesn't develop advanced UIs with their frameworks. WPF is still well-used by Windows and other Microsoft utilities like Windows Terminal. They are just stupidly abandoning their built up bases for silly industry fads.
They jumped into tablet / touchscreen / hybrid-mobile-desktop bandwagon in late-2010s and tried to force WinUI as an UWP-only feature. It resulted in low adoption. They didn't adopt WPF to have same theming.
When WinUI2 failed, they tried to make modern C++ a reality and tried to remove UWP restrictions which is a good decision. However they diverted quite a bit resources into AI slop generation now and WinUI3 just languishes.
Same for MAUI. They tried to get into multi-platform, multi-device framework as a way to generate leads into Microsoft ecosystem.
They try to use various frameworks and UI stuff to get people hooked into the ecosystem and find ways to upgrade them into Azure and M365 customers. It is meaningless and tiring. All of those could be only WPF.
It is like Google and its many Bazel-like build systems (but not full Bazel) for each of Chrome, Fuschia and Android.
MAUI was horrible when I tried using it about a year ago, tons of bugs, pretty iffy comms/support from the MAUI team as to timelines when things might get fixed, etc.
Eventually dumped it and moved to Kotlin Compose/Multiplatform, which is just so much better at achieving a similar goal (though, obviously, without being part of the .NET ecosystem).
microsofts own stuff never seems to be what gets momentum. there's a strong aftermarket for better ways like back in the borland era bcb and delphi, the more things change the more they stay the same!
Thank you. I was afraid someone would end up saying employers don't give a damn about personal projects. I've came across this settlement before as employer only care what you did in your day job.
As for stack, I can definitely see demand for Python/node.js for early startup companies where I prefer to be in as of now and move away from enterprise.
On the certification side, my genuine reason is to have structured approach to learn some topic. What is your opinion on picking up Azure and learning the "admin" side of it, like managing group policies or enabling rbac? I might be impatient but this part of azure solution architect perquisites really makes me loose interest in cloud route. Like the fun stuff comes later.
On certification, I can also try to pursue machine learning route by Deeplearning ai or Google.
MLOps does sound something exciting to do since I enjoy tinkering with Linux and deployment.
I'll be honest, I don't want to get myself more onto .net framework. .net core is still fine but what I've seen, there is way more demand for Python/node.js
I don’t have anything against Azure as a servíce. But Azure is mostly popular among “the enterprise” and government. Microsoft shops naturally tend to prefer Azure. If you want to work with startups and get out of “Big Enterprise”, Azure isn’t what you should be focused on.
On the other hand, going from .Net Framework to .NET Core + Azure may be the path of least resistance to get into a
modern ecosystem. Despite what you read on HN, most developers work in boring old enterprises.
As far as startups, ageism is real. But it is more nuanced. They will hire older developers. But only in if you have the experience “you should”. They offer me jobs when they want someone strategic, and knows the modern tech scene.
But understand, the issue with focusing on “in demand” skills is that everyone is doing it and it’s almost impossible to stand out - it’s a shit show out here right now. You’ll have a lot less competition in the Microsoft ecosystem and it will be a lot easier to convincingly tell a tale about using modern MS technologies.
I am not however discouraging you from leaning deep into the JS ecosystem, its table stakes and has been for 20 years.
I genuinely want to benchmark myself with fellow peers how on long they take to consume text book. I am not that into fiction or short fiction books - I generally loose interest.
How long would technical books take you to complete, say you have to read Effective Java 3rd Edition
I haven't read a technical book in a long time. I read "You Don't know JS," which is very very short, in about a week. I read "Cracking the Coding Interview" in 3 weeks.
As for non fiction, I read "Anarchist Communism" (fairly short) in a week, "Delivered from Distraction" in a month, "Masters of Doom" in about 2 weeks.
It depends on my interest more than anything. I obliterated the entire Robin Hobb "Assassin's" series, 12 books, in just about two months.
I can assure you a lot of veteran devs in my company who had major impact were let ago because of off shoring to India and Poland. My employer is a very well known financial institution, think Blackrock, Bloomberg etc.
Public press paints picture of the firm laying off due to adoption of AI. Yet, I can assure you without doubt its not because of it but due to continued offshore.
I think we are at the brink of opening another large campus in India, not sure which city. Oh another thing is these full time position in NA were eliminated and replaced with employees from TCS.
My own boss is against it but its coming from higher ups. I have no idea what's driving them. We made billion in net revenues last year and we are actually part of the org that is in 'cash cow' category - upper management literally said this too in our town halls
Exactly, you're on the hook to profit-max for the next quarterly earnings report. If that may negatively impact subsequent quarters, not your problems, that's to be resolved when profit-maxing those quarters
No, you're not "on the hook". You're definitely motivated to increase profits - most senior leadership's compensation will be heavily equity weighted, and net profit growth tends to result in share price growth - but there's no hard rule saying they have to purse short term profits at all costs. They might be pressured to by shareholders but they're not obligated to.
Consider companies like Amazon, Tesla, who (initially, for many years) prioritised long term growth over short term profits. Look at how long they lost money for. You do need a CEO who actually has some vision.
One might argue sacrificing long term gain and stability for short term term profits are indeed a violation of fiduciary responsibility. Why it isn’t being argued I don’t know but we all know delaying gratification tends to result in higher net in the long run.
Yeah, same here except in Retail. All the hiring is in the Indian campus where they can pay junior devs 11/hr compared to the 80/hr billed by Kforce, InfoSys, et. al.
Let them offshore if that’s what’s going to happen. And then let the stats show that American employment in certain sectors is tanking, and let the electorate decide what to do about that. Basically, I think this is a multi-round game rather than a one or two round game.
how are you managing to run and work between 8 concurrent agents to get through the ask in hand?
I keep seeing git workspace tree + agent combo being mentioned
reply