guzik

Gee, today is my birthday (36). I've never managed to get anything I've built to the HN front page. Always wondered if that means my ideas just weren’t that interesting, or if it's just the luck of the draw.

*edited my original comment without mentioning my project*

let me piggyback on this thread because maybe there’s a kind soul here who knows the answer. I set up some simple cron jobs in Claude Code that run every few hours. I am on the Pro plan and unfortunately, I cant fully solve this one problem: I keep getting the error:

Invalid API key · Please run /login

I know there are a lot of github issues about this but most of them dont seem to directly apply. The weird thing is: when I run the exact same command manually in my terminal, it works fine. When run via cron, it fails. I suspect it has something to do with how Claude CLI stores the session token or environment variables, since cron doesn’t run as a full login shell. I already tried wrapping the command in bash -l -c "..." which helps in some cases, but not consistently. Has anyone here figured out a clean, reliable way to run Claude CLI from cron without having to manually relogin every 5 hours or so? Or is it only restricted for paid anthropic API Key?

just 1hr ago (1 AM local time) I saw 'your app is live on app store' notification on my phone and eagerly launched it... only to have it crash instantly. After a debug session I discovered an obscure bug in tflite library that only shows up in release builds. 20 minutes ago I pushed a hotfix with an expedited App Review request, hoping to spare as many users as possible from that crash. I can't wrap my head around how the appstore review missed it, especially after rejecting our last build 4 times over a barely legible location-permission alert description.

That said, I built my first mobile app 15 years ago, and to this day, building for mobile remains the most frustrating part of my programming life.

we've built a wrapper around Apple Health to simplify data extraction and synchronization. If you think this might be useful, feel free to drop me a message.

Those who made it to the Birmingham show were very lucky

Had an interesting encounter recently. Was coming back from a festival and got talking to this woman who mentioned she has to be careful about her health. Turns out she has a pacemaker. Since I'm in the medical device space (much lower risk category though), I was curious about her experience. What struck me was how much she knew about her device like the technical specs, failure modes, battery life, everything. Makes total sense when you think about it (if your life depends on a piece of tech sitting in your chest, you'd probably want to understand it inside out too).

Are you comfortable with chest straps?

I might be missing something, but is there any _practical_ value in this line of research beyond academic curiosity? I've stumbled across this article a few times already, and still can't quite find the real-world application where you'd want to identify someone just by their breathing pattern (especially considering that from the article that "you need to be equipped with a nasal cannula"). Maybe I'm being dense?

glad you mentioned xbar, and your comment is ranked high (so maybe someone can help me out). I've been a long-time user of xbar, but recently it stopped showing up in the menubar on macOS 15.2. Even a simple script with just print doesn't display anything. Anyone else running into this?

btw. I'm starting to think xbar might be abandoned - it hasn’t seen updates in a while.

This is hilarious. I'm also confused about whether they released it or not because the results are underwhelming.

EDIT: Ok it works in Sora, and my jaw dropped

I'd add pitching here: https://aitinkerers.org/. There are lots of such events in major cities. I've been to a couple, and they were awesome (quick 5-minute demo formula)

LD is becoming popular again? I feel like I’m seeing more and more threads about it lately. Even my friends are getting back into practicing LD after abandoning it because of how hard it was to achieve. I think I’ve had a lucid dream twice in my life-each lasting about 4–5 seconds - and honestly, those were some of the most exciting seconds I’ve ever experienced. It’s like being a god, even if just for a moment. The problem? To get there, I had to grind for almost a year—writing down my dreams, doing reality checks etc. Too much effort.

Where are the images on your blog?

> does your organization's S3 costs ever become a problem

A bit. We moved to a stricter data retention policy.

Pebble is still the best and the only watch I've ever owned. Nothing has matched its simplicity and functionality for me.

Gumroad was dishonest to us on so many levels. Fortunately, it was around the time Stripe became available in our country, so we could move our 'toys' to a new payment processor (I know Gumroad is also a marketplace, but that was irrelevant for us).

I'd give anything to move our apps to the web. We're in the medical field (our apps connect to medical devices over Bluetooth), so publishing on Google Play is like a Kafkaesque fever dream. Two days ago they tell us our app fits into categories like:

Activity and Fitness

Nutrition and Weight Management

Sleep Management

Medical Device Apps

So we gotta update our policies, but today they said something like: "Actually, jk, you don't qualify as a Medical Device App anymore - app update rejected."

Meanwhile, Apple (who used to be the actual nightmare) has somehow turned into the reasonable one. They approve apps in minutes now. MINUTES.

It's like Google saw Apple's approval process from 10 years ago and thought, "Let's do that, but make it a circus." At this point, I dream of ditching native apps entirely. Too bad Bluetooth on the web is still a bit... fragile.

Clearly, in Cupertino, 'enhancing user experiences' without consent is the top priority.

Imagine this: Newton formalized calculus in 1671, yet the first (known) visualization of a continuous was drawn in 1669.

Don’t even get me started. We’re in wearable electronics, and somehow, we have to compete with devices priced under $15.

I know there are people who say, “Why pay more?” but honestly, at that price, you’re not just missing out on quality materials that have been thoroughly tested. You’re also missing proper support, reliable updates, and the peace of mind that comes with a product designed to actually last.

Ok, but I think most popular health apps and wearables offer access to data via an API (allowing you to pull the data periodically) or through a CSV export. At Aidlab, if we didn’t provide this, most of our clients would quickly find an alternative.

By the way, which of the apps you’ve used have been discontinued

Do you think they might start monetizing, for example, by prioritizing certain companies' products over others? For now, we are getting some traffic from chatgpt whenever someone asks for sleep and heart rate monitors, which is nice.

_Somebody Has Already Figured It Out and Won’t Share the Results with You_

For easy questions, sure, it’s way simpler to find answers now. But relatively speaking, if building a raft in 3000 BC was super hard, today’s equivalent might be asking how to build a Falcon 9 from scratch. You can’t. It’s a secret SpaceX keeps to itself. Not to mention the Biotech/Pharma industry and their advancements on drugs discovery. They know how, but you don’t, and you probably never will.

It only affects 3rd party analytics, so it should not.

I've been using Thunderbird as my default email client for about 15 years. Despite multiple attempts to switch to other clients, I always end up coming back to it. I’m not sure if it’s just out of habit or if it’s genuinely that good. One thing I really wish for is for "Send Later" to be a built-in feature instead of relying on plugins, which I find unintuitive for such a simple function. Either way, big kudos to the Thunderbird team!

I’m not sure if anyone else has really tried, but I’ve tested it a few times and never hit meaningful results.

1) I tried using it for QA for my SaaS but agent failed multiple times to fill out a simple form, ending with it saying the task was successfully completed.

2) It couldn’t scrape contact information from a website where the details weren’t even that hidden.

3) I also tried sending a message on Discord, but it refused, saying it couldn’t do so on someone else's behalf.

I mean, I’m excited for what the future holds, but right now, it’s not even in beta.

Weird submission. When I open the link I read that the tool is deprecated.

There are also plugins that nicely integrate with ESLint to do this.

Unsurprisingly, I don’t see many hardware projects here.

If anyone’s looking to share their story, I’d recommend IndieHackers, but for hardware-focused startups/projects, check out howtoware.co. (I’m not affiliated with the site’s owner in any way, but I really like the concept and want to help spread the word to my fellow hardwarers).

> - Once your core product is built, its worthwhile spending some time automating the heck out of everything. This will save a TON of time in the near future.

Interesting. Anything you've automated successfully that you can share? I've heard so many times that you should hold off on automating too early because constant pivoting and refining can end up making you spend more time fixing the automation than actually doing the work itself, so I kinda paused on it. I can see how it would make a big impact on my marketing outreach, which I'm doing manually right now with not much success.