Things I've built in the last month or so:
* A personal music player for Apple Music with an embedded agent loop, so I can say things like "I'm going to work in the wood shop, play woodworking-appropriate music and keep the playlist full and non-repeating indefinitely".
* A calorie/protein tracker that takes 1-liner summaries of meals, resolves them to macros with GPT5 calls, and tracks them in SQLite.
* A macOS menubar combined Apple TV and Roku TV remote control.
* A calculator front-end to SageMath that renders Latex math notation and has point/click shortcut affordances for linear algebra and MVC.
* A "self driving wiki" macOS desktop llmwiki that imports and digests PDFs.
* A menubar tracker for the little temperature sensors I have scattered all throughout the house so I can click my menubar and instantly see the temps in the different rooms of the house.
* A layout tool for designing picture frames to get cut lists and rabbet depths and stuff.
That's just the desktop stuff. These applications are generally better than what I'd get from the App Store. The calorie tracker is one of the better-designed apps on my dock.
I feel like I'm not doing this enough yet, it hasn't fully sunk in, and I'm going to be building a lot more stuff when it does sink in. I haven't yet gotten to the point of writing my own note-taking application (still Bear.app for that), but how long can that last?
As a new father, just using so many apps sounds exhausting, let alone building and maintaining them. I would have been very excited about these maybe 10-15 years ago. Now I just want the ability to do things with my phone when my hands are tied up holding the baby.
As a father of two young children, LLMs have been kind of amazing. I’ve historically been someone who has project ideas and has fun building them. But for a few years now I just haven’t had the headspace for it — all I’ve wanted to do in the evening is relax for a bit.
In the past month I’ve completed a home dashboard idea I’ve had for years, I’ve built a house management system, a travel planning tool, a guitar chord annotator… it’s been very liberating. I’m not doing the coding, which was fun, but I am doing all of the system design and iteration, which I also really enjoy.
I’m building a video game for our abernic device with my son (no game engine experience, just vibing the game engine ourselves). We’re having a blast. We play up to our progress right before bedtime stories.
The other day I refused to fire it up for him and he had a full meltdown. I told my wife I had never seen product market fit like that before.
It is a little harder with something like a game to maintain motivation and go for perfection when you know your audience is an audience of one but it’s also a relief to just worry about the things he cares about (trains and firetrucks).
I also miss there being something intellectually challenging to a side project but that probably means I just need to learn a real game engine after this.
Similarly, I have giant list of ideas from the last 15 years that I’m now easily able to try or build out in ways I couldn’t before
That's an odd response, because fundamentally all I'm doing is making my computer work the way I want it to. It's the opposite of exhausting to me.
Nothing against what you are doing. That’s awesome and exactly the kind of thing I would nerd out about. I was just commenting on my current state of not having the mental bandwidth to do these things.
Most people want the convenience of things just working at the tap or click of a button. LLMs are not going to change that.
I was talking to my wife about this a little bit, that the promise of computers has never really been realized in the sense that we don't put power in the hands of the end user.
Like one of the most obvious things that a computer can do is run a task periodically at a time to go maybe fetch information and process it. And yet, the average computer user using any commercial operating system does not understand how to set up a recurring task. Operating systems don't really advertise it or talk about it as a feature. I think this is a good example of the disempowerment of end users by software companies.
LLMs have the potential to change this by allowing people to simply ask their LLM to invoke the right incantation to schedule the task or produce the program. That seems like a great change. I'd like to see more power in the hands of the end users rather than less.
For people that don't want to put in the effort, of course, this will change nothing. But for those that are willing to explore a little bit, I think it can be a big improvement over the software ecosystem that's emerged in the era of the app store.
The average user can and already does that. If they want to regularly fetch their emails and process them, they will install outlook or use the Gmail web app. If they want to regularly fetch and install updates for their games they will install steam.
The "run a task periodically" isn't the difficult part, it's processing the data that is.
And to be honest I don't think llms will empower that many people, the vast majority don't want to put the effort required (and it's definitely more effort than writing a one-shot prompt)
Those are just built in functions in the apps. I think we are more talking about custom automations like ”I want you to download the images on this site everyday and put them in this folder and send me a message on telegram with any new photos” or ”I wanna check this website for searches for X and when a match then message me”. An example I have seen is non-tech people before logging in to a saas app and manually copying numbers to a spreadsheet every single day forever. But with LLM:s they automated it to just work automatically in the background
Exactly. The users of e.g. autohotkey are in a vanishing minority. It's the kind of stuff I'd like to see advertised more in the OS, as though a product manager had on OKR to move that metric. Apps like Apple Music that emphasize consumption (and subscription revenue) get that treatment, but empowering apps that teach automation do not, since there's no (short term) business there.
I agree. It won't empower them because they've been trained to think they have no power.
I wasn't really talking about fetching email.
Most of the apps I've built take the form of "something I was doing with a long-running note in Notes.app that is now much simpler to deal with because there's a purpose-built program for it". That's what computer programs, like, are?
> maintaining them
In the agentic era, there is no need for maintenance - each artifact is immutable and just-in-time. In case you need a new version, spin up a new artifact.
I’m only half-joking.
Tell me about it. Holding a 4mo old right now, nap trapped :-) Speaking of which - Claude Code wrote me a web proxy for HN that rewrites CSS for the dark mode - precisely for this situation. Written and deployed with Claude Code remote control too, also while nap trapped.
Haha amazing I’m doing the same thing
Wow okay I need to try this out.
Make yourself a dashboard that pipes everything you want into the area reachable by one thumb. :)
Or with voice. Or even face movement control, like via the camera.
> A personal music player for Apple Music with an embedded agent loop, so I can say things like "I'm going to work in the wood shop, play woodworking-appropriate music and keep the playlist full and non-repeating indefinitely".
how is AI (or a human for that matter) supposed to know what “woodworking-appropriate music is”? I had the same thoughts when spotify released their “let AI make a playlist for you”. Like, we already have recommender algorithms, why do we need to get a LLM involved?Maybe I just don’t understand woodworking and the connection it has to music though (I’m also the kind of person who’s had the same playlist for like the past 4 years)
But, I do want to say, personal app built just the way you like it, an alternative front end to a closed source ecosystem even, is super cool, and has been one of the things that AI coding has really enabled.
I’ve done a few project where I just give claude a har file and tell it to reverse engineer a cli from some web interface and it’s so powerful to be able to do stuff like that now
> how is AI (or a human for that matter) supposed to know what “woodworking-appropriate music is”?
It doesn't, but you'll believe it is because that's what you asked for. It's not like you have a solid test for what it is yourself
Guaranteed it would be unable to avoid searching for and selecting songs that specifically mention wood, woodworking, the shop, etc. It's like when you tell it to write something and give it a "vibe" or "mood" word wishing it could just give tone to the generated text, but instead it incorporates it literally.
Now, see, when AI does something like this, we call that "hallucination".
Thomas explained what he means in a sibling comment an hour before yours.
Because I played a bunch of music while doing woodworking stuff and asked for more like that. Things like this are easy with a simple agent loop. It's not picking random music out of the sky; it's working from my Apple Music library (and then expanding on it from the Apple Music catalog).
>I played a bunch of music while doing woodworking stuff and asked for more like that.
I would probably think about doing a little workspace sharing and ask for more live musicians to come in and do some serious woodshedding themselves ;)
I know some that are getting that buzz-saw sound out of their instruments fairly repetitively already, and there's no place to go but up :)
I'm a terrible woodworker (that's what makes it fun) and if I had a musician in our basement shop there is a very high likelihood I would kill them, or at least seriously fuck up their instrument, when I manage to shoot a wooden spear across the room because of miter saw kickback.
> Things like this are easy with a simple agent loop.
To be fair, they’re also easy (trivial, even) without one. It’s not like “play more music like this” is something we ever needed AI for.
> It’s not like “play more music like this” is something we ever needed AI for.
Borderline off-topic, but I love how fast the goal post moves. A decade or two ago "Similar music to this" was definitively in the realm of AI/ML, now it's some full-stack dev's afternoon to implement.
I don‘t think someone disputes that this is in the realm of AI/ML. I just think its questionable if this kind of usecase is genuinely something were an LLM is necessary or outshines the more traditional approaches
What are you even talking about? No, of course music recommendation systems had nothing to do with LLMs two decades ago (which is obviously the crux of the discussion), that is insane. And no one has said anything about a “full-stack dev's afternoon”, I’ll kindly ask that you don’t straw man. No goal posts are moving, that fallacy does not mean “argument I don’t like or understand”. What happened to good faith discussion? It’s ever more common on HN that someone will just invent some argument in their head, attack that, then act like they are the arbiter of reason when they couldn’t even understand the argument. That’s not interesting. Engage with what was said.
> What are you even talking about? No, of course music recommendation systems had nothing to do with LLMs two decades ago (which is obviously the crux of the discussion), that is insane.
Of course not, but just some years ago, when people talked about "AI" in ML circles, they weren't exclusively talking about LLMs, which you seemed to have automatically and exclusively associated "AI" with. It's much larger than just LLMs, although the ecosystem is flooded with LLMs at the moment. I thought this very movement, where tons of people, not just you, have this "AI === LLMs" association, when again, it used to not be so.
So, in other words, you read what one person (you’re not talking to some vague notion of “people” in the past) wrote, associated it with other arguments you’ve seen (and quite possibly misinterpreted) before and decided to lump everything together and answer those different arguments and be kind of snarky about it.
That attitude is exactly why online (and offline) discourse has gone in the gutter. Instead of addressing what was said and understanding nuance, you instead took the comment, labeled it in a preconceived box in your head, then attacked that group instead of engaging in good faith with the person you were talking to.
And by the way, I’m very much aware of the difference between AI and LLM, but whether you like it or not the terms are conflated today and being pedantic about it is a losing battle. In this case it was patently obvious what was meant. You’re (presumably) not an automaton and are thus (presumably) capable of understanding the difference, meaning you are definitely able to choose to steel man instead of straw manning.
> than attacked that group instead
What on earth are you on about?
Take a breath, I'm not arguing for/against anything at all, I conceded myself it's "borderline off-topic", you don't have to take it as I'm trying to challenge anyone's worldview or anything, it's basically an offhand observation.
> man instead of straw manning.
You may not actually understand what a "straw man" is, as I haven't even attempted to refute or engage in any argumentation at all.
Anyway, clearly a fruitless conversation to continue, I'm not sure who you're arguing against/for here, I didn't intended to even try to argue against/for anything here, so I'm just gonna peace out. Hope you enjoy the rest of your Saturday as much as I've enjoyed mine so far!
Come on, if you respond to a comment then expand on it to explain why you made it, hiding behind it being “borderline off-topic” and an “offhand observation” feels disingenuous. Can you understand that?
Here’s a silly example: Imagine you had made a comment about something different on a different thread. Then someone responds to you by saying “borderline off-topic, but I find some comments on HN about this subject to be really stupid”. Would you not be justified in thinking that the comment responding to yours was a response to you?
Maybe you really did mean nothing by it, but are you able to understand how every choice you made contradicts that, even if inadvertently?
> Anyway, clearly a fruitless conversation to continue
Are you also able to understand how this type of high-ground dismissal makes it worse?
> Hope you enjoy the rest of your Saturday
Thank you. You too. Genuinely.
Oh, you two!
The true power of LLM powered interfaces is the malleability of the whole thing. Sure you can write traditional software to handle “more like this”, but how hard would it be to handle “more like this but uptempo”, “more like this but not the same artist”, ad infinitum?
I have a bunch of skills Ive created for random work that have “predefined interfaces” like a single url. I very frequently trigger these skills with freetext prompts instead and get all the intelligence of the skill while bypassing what it was actually designed for. This aspect of malleable API is extremely powerful and is probably the only reason I keep AI in the loop of some of my custom software vs removing it and avoiding the cost/nondeterminism.
Recommendation algorithms are, frankly, weird as a user. Something watches your tiny interactions and, with zero concept of anything happening, gets surprisingly good at predicting your next interactions.
It’d be like eating and having an alien with no concept of food or taste staring at how much your facial muscles change and making notes. Then recommending places to you based on the other people they’ve been studying.
> how is AI (or a human for that matter) supposed to know what “woodworking-appropriate music is”?
Well someone is going to be concentrating, but it’s a long period, with loud activity. maybe you’d guess at decently active music, not relying on careful complex listening, probably fairly steady rather than sudden dramatic changes.
Maybe it could save previous playlists and what you’ve said about them, or ask questions like “so like, really calm or techno or what?”. Adding just a bit of common sense makes an enormous difference with this kind of thing.
They are weird. But stuff like this solves one of my great challenges in life, which is that I've got many tens of thousands of songs in my library but only manage to hear tiny repetitive epicycles of them. I don't want a tight rotation of the same 200 songs, and I don't literally want to randomly walk my whole library, and I don't want to allocate lots of time curating exactly what I'm going to listen to. LLMs are great at this.
(Actually I totally do want to allocate lots of time curating exactly what I'm going to listen to, but laying out playlists is its own hobby, and woodworking time is for making wood shoot across my basement when I pinch the saw blade, not for making mixtapes.)
Oh I mean LLMs are the opposite, they’re not the same blind optimisation thing weirdly watching you with no core understanding. They’re great for this kind of thing.
Woodworking recommendations shall include:
“If I Had a Hammer”
“Sledgehammer”—Peter Gabriel
Anything by The Carpenters
Any works by Woody Guthrie
Hymns about Jesus and St. Joseph
Lumber Capital Log Yard
Anything by Nine Inch Nails
Anything by the Sandpipers
“Turn, Turn, Turn”
“Dust in the Wind” by Kansas
Anything by Dusty Springfield
Anything by MC Hammer
Original Motion Picture Soundtrack, “Saw” horror franchise
Anything by Level 42
https://youtu.be/dQw4w9WgXcQ?si=sTUKhRXXiylavwIm
Anything by 1-hit wonder The Nails and/or powerviolence band NAILS
“Morning Wood” by Beavis & Butt-Head
“The Lumberjack Song” by Monty Python
> we already have recommender algorithms, why do we need to get a LLM involved
and we had autocomplete for coding (IntelliSense), why do we need to get LLM involved?
LLMs are at a whole different level than recommender algorithms
tried it with GPT:
> A steady mix of instrumental grooves, classic rock, and bluegrass—styles commonly favored in woodshops.
Green Onions - Booker T. & the M.G.’s
Up Around the Bend - Creedence Clearwater Revival
Jessica - The Allman Brothers Band
Long Train Runnin - The Doobie Brothers
Honey on My Tongue - Steep Canyon Rangers
2 of those acts are in my woodworking rotation.
> I feel like I'm not doing this enough yet, it hasn't fully sunk in, and I'm going to be building a lot more stuff when it does sink in. I haven't yet gotten to the point of writing my own note-taking application (still Bear.app for that), but how long can that last?
Indeed.
I'm not sure when, but at some point after I saw Claude Code, I realised what I wanted for my next OS was basically an LCARS-themed* thing, where the "desktop" is a TUI and whenever I needed an app I didn't already have, I'd ask an LLM to write it for me instead of downloading it (benefit: the app would never have an update pending (which might break my flow) that I had to install before use, unless I'd specifically asked for a new feature to be written); and if I wanted to browse something on the web, the browser would use something like an LLM to strip out everything that wasn't content.
LLMs aren't quite good enough for that yet. They're interesting, I've also made a bunch of apps this way that was an easier experience than even figuring out which offering on the app store was right for me**, but still, the LLMs are not quite good enough; I'm pining for the OS equivalent of a car with a self-driving mode so good there's no steering wheel and the windows have black-out shutters so you can sleep while it drives you and your insurance is fine with this.
* because I'm that kind of nerd
** e.g. rhythm action game that takes midi files for the songs and listens for a midi keyboard for your input: https://benwheatley.github.io/Piano-Trainer/piano-trainer.ht...
How are you counting calories from a 1-line description of your meal? Surely that’s not even close to accurate.
If it includes all ingredients, cooking method, and weight in exact grams, it’s probably reasonably accurate.
For basically anything but chicken I give exact gram weights of proteins and of any fats or carbs. I generally don't give it any precision about vegetables that won't drive protein, fat, or carbs. Most meal components aren't going to meaningfully change the macro breakdown except for fiber, which I don't care enough about to track.
You should care more about fiber than any of the others if health is your goal.
I'll make sure and let my GP know he was wrong.
Snark aside, your GP took one module on nutrition in medical school, years ago.
Doctors are not experts in nutrition.
Most modern western diets lack fibre the most so eating more fibre is a sensible recommendation for basically everyone.
[flagged]
People are trying to tell you something that can literally save your life. Colon cancer rates are increasing at a breakneck pace.
Chill with the snark for a second and realize you may not be the smartest person in every room.
This is going to be a funny thread to show my doctor. I think something that isn't getting through to people is: I'm not taking The Internet's advice on this stuff; I hired a person to handle these questions for me. Highly recommended!
I quite literally, right at this moment, am sitting next to a friend who is a doctor with a focus on nutrition and they are saying that whoever told you that fiber doesn’t matter is going against almost all of the modern understanding of health science and research.
His advice is literally “He needs to just Google it.”
They didn’t say fibre didnt matter. They disagreed they should care more about fibre than any other.
> except for fiber, which I don't care enough about to track.
I can't recommend enough that you go build your own tracker that graphs your fiber intake, shows trends, and alerts if you're under threshold. It's not hard! I do that for protein, and it's been very valuable.
It's too easy just to drink the glass of pysllium husk and not obsess over data entry and trend analysis. You may even learn that your poops are the alerts.
All that snark and you're still wrong, fiber is very important OP is correct
A typical (in fact, the modal) line from the database, which goes back to April, is "oyakadon 1/2c rice, 2 thighs, 1/2 breast, 2 eggs, bunch broccolini, std oyakadon broth", and the model works out for itself that these are average-sized thighs and breast (boneless and skinless, because it's oyakadon). "std oyakadon broth" is a reference to a "house recipe", of which there's a side database, each costed by GPT5. There's a low/med/high adjuster on every entry; sometimes the thighs are bigger or smaller, for instance.
I've:
* Checked these against all three major frontier models and gotten the same macros breakdown within relatively tight bands
* Manually researched a couple of the most common recipes
* Run this system over many months against my most likely TDEE and reconciled against scale weight loss (which has been profound).
My confidence in frontier model's ability to generate "good enough, probably more than good enough" from this level of meal description is quite high.
Key simplifying assumption (for the model): I'm cooking most of my meals. I cook. I virtually never order out or eat packaged food of any sort. This would be a harder problem and the answers less reliable if I was giving it, like, "1/2 JP Graziano muffaletta, extra hot giardiniera".
You can actually get just as good results for packaged food if you just allow the agent to use a web search tool & keep a growing local database of "known food items". I did the same for a while, but it was quite expensive (this was a while back with less powerful models and required a bit of orchestration to make it reliable)
Presumably it would Just Work if I wrote "1/2 bag cool ranch doritos", which is one of the virtues of this input/processing system.
perhaps using ai.
uh oh
@tptacek (and others who have their house full of sensors and toys), check out https://httpstate.com.
Your feedback as a builder is greatly appreciated :).
I feel on the opposite side of a divide when I see these kinds of comments here about people creating bespoke software to address deficiencies they perceive in their daily lives. There's an engineer mindset that tools can solve any and every problem, but do they? Are you quantifiably any fitter than you were before? Better bloodwork, BMI, body fat percentage, anything? Are you creating better wooden goods? More wooden goods? If you track mood in any way, are you at least consistently happier while working because of the auto-generated playlists? Have you gotten any better at math thanks to the pdfs?
When I reflect on my own life and think about why I'm not personally adopting LLMs for anything, it occurs to me that I just don't feel like I have any problems like this to address. I was already eating appropriately, already working out appropriately, already happy with the music I listen to, already happy with the way I learn and practice math. I could make a billion personalized apps, but I was already a two-time state champion in cross-country and track who has never been overweight, got perfect scores on the SAT and GRE, won regional academic olympics in math, all before mobile apps existed and before I ever owned a personal computer of any kind. That's not to say I can't do even better than I did as a teenager in the 90s and I try to maintain some proficiency in these things into middle age, but it never occurs to me that I need or should need software to help me with this. I have no clue what the temperature is right now of any room in my house, but I know they're all in a comfortable range that isn't hurting any machinery or animal and that seems good enough to me.
I see all these enthusiastic booster posts and comments about all the tools you can make, but I don't feel like what is missing from my life is I don't have enough tools. I wasn't sitting around prior to GPT-3 thinking to myself "you know what would improve my life? More software." There are deficiencies and problems to address for sure, but it's mostly issues of prioritization and motivation. I have fallen off the wagon a bit this year and don't exercise nearly as much as I should. I don't talk to my wife enough. I don't talk to other friends at all. I don't go outside enough. But those aren't problems to address with software, either. If anything, if I spent more time making software, even with LLMs, that would be more time not going outside, talking to my wife, or working out. No matter how much raw output I could get computers to output, more time spent fiddling with computers is not going to improve my life. I shouldn't even be here right now, reading and commenting on Hacker News. The sun will be up and it will be 90 degrees fahrenheit within an hour of that. This is my last chance of the day to go run without feeling extremely uncomfortable while doing so.
This isn't a "oh my God look at me I'm so much better than you" comment or at least that's not what I'm trying to get across. But I'd encourage the original poster here and you as well to be outcome-oriented. My weight isn't in a warning zone. This guy is posting stats, so I am leaner than he is and my marathon time is faster, all without the aid of any tracking apps. You're a public figure, so I know you're in the same time zone as me, which means you were posting your own comment at midnight last night, midnight on a Friday night reading Hacker News. My sleep isn't perfect, but I'm not reading Hacker News at midnight on a Friday. Are all these apps actually making your life better?
> There's an engineer mindset that tools can solve any and every problem, but do they?
No. These people just spend more time fiddling with their computers instead of doing the parts of life that it supposedly frees them up to enjoy. It's IoT lightswitch automations all over again. There's no reason anyone would ever need to calculate their smoothie according to their Strava data and the positions of the planets on that day. It makes far more sense to spend more time knowing your body and learning to treat it well than trying to create an app that will do it for you.
My running joke for the past couple years has been "there's nothing a software developer hates more than computer science". But I guess there was one too many words in that sentence!
As someone fiddling with computers trying to solve "every problem," the only actual solution for me is making tools to assist with being aware of the state of your body. Overly simple example, but give someone a heartrate monitor for the first time and they won't be able to guess their current bpm all that accurately. If they pay a little attention to it for a month, you'll get much more closer guesses for instantaneous rate. If it becomes easier to be aware of your current state, you get a more accurate model, and can start to predict basic cycles etc. Eventually, without the tools.
I am wildly, drastically healthier today than I was when I started this in April. This is something I should have been tracking and measuring a long time ago.
i get what you're saying. but i still feel it necessary to say you sound like a twat.
I've had a personal health dashboard tracking my metrics for the past four years. Yes I am quantifiably healthier than I was.
You've climbed a certain mountain, and you're strolling around the summit on flat ground.
You're looking down on the people at the bottom or middle of the mountain, struggling up hill to reach the peak, and wondering "why are they going through all this effort? Why don't they just stroll around on flat ground like I am?".
They're still building the systems that you already have in place. It's far easier to use a system you already have than to build one from scratch. Maybe not with physical or academic health for you, but I'm sure there's some areas of life that you're lacking in that could use improvements. If you can't think of any, I can give you a hint: empathy. If you try to work on it, you might have to do things like set reminders to call friends on their birthdays, or just to check in. You might have to force yourself outside to chat with that neighbor that you don't exactly like and find some common ground. You might have to set a goal of volunteering a certain amount of time at a homeless shelter, and track those hours.
Meanwhile, there are people who already have buffed up their empathy their entire life, and do all these things(and more) naturally.
Should they look at you and wonder why empathy is so hard for you, and tell you your efforts to consciously improve this part of your life is a waste, since they were able to achieve higher levels of empathy without doing any of those things ?
> They're still building the systems that you already have in place.
Bull. They're doing highly sophisticated procrastination. I'll never become a great photographer by vibe coding the perfect tool for my workflow that selects, ranks and edits the 100 photos I took this week, or one that tells me where to stand and when to click the shutter. I have to be out shooting thousands of photos every week and laboring over the editing myself. You get good at what you do, and if you're spending your time making these one-off disposable apps instead of doing the thing they ostensibly support, a few years from now that's what you're going to be better at.
I think the whole point of OPs post is that they didn't spend a bunch of time coding the apps, since a LLM did it for them with a fairly simple prompt.
> I have to be out shooting thousands of photos every week and laboring over the editing
Why? Ansel adams would only shoot 20 photos a week on average. Why are you procrastinating by spending so much time with the camera instead of just having naturally good taste like him?
What if by using custom software I’m not pissed off by my tasks anymore, or no longer surveilled while doing them?
What if I just like making my own stuff?
Surely that’s plenty.
This sounds like such a security risk. It just takes one bad library. Your phone and desktop are not a sandbox.
Don't think it has to be, if you are mindful of it. Ai uses fewer dependencies usually, as it doesn't mind rewriting things, and you can easily add scans and sanity checks to your sdlc
These are all simple SwiftUI applications with minimal dependencies. My phone, by the way, definitely is a sandbox.
Lmao you're talking to someone who is an accomplished security engineer.