Live data from Hacker News

Ask HN: What non-AI products are you working on?

news.ycombinator.com

421–430 of 544 posts

Re: Ask HN: What non-AI products are you working on?

#421

I am productionizing a golang project which I have working ... it parses an input image at the pixel level to synthesize its audio equivalent ... inspired by a 3Blue1Brown video on Hilbert Curves ... as it traverses the entire image pixel by pixel it assigns to each pixel an audio frequency which it increments for next pixel ... in doing so it collapses the 2D image into a 1D line of pixels ... then it engages audio…

That sounds fascinating! I'd love to see the results when it's ready.

Re: Ask HN: What non-AI products are you working on?

#422
post #382
post #95

Working on Lungy: https://www.lungy.app It's an interactive breathing app (responds when you breathe). Initially for stress & anxiety, we're developing a medical device (SaMD) for asthma + COPD.

Reminds me of Breathscape [1], a former colleague's biofeedback-driven generative audio app. [1] https://www.breathscape.com

Thanks for sending! Yep, I have tried it before - it's cool and definitely in the same ballpark.

Re: Ask HN: What non-AI products are you working on?

#423

Earlier quoted context omitted.

What these reptilians do is only think about themselves. If you're invited for dinner it doesn't mean that somebody owes you a meal. It means somebody wanted to make a nice gesture towards you and get to know you more intimately, perhaps to discuss important things. You eat something before you go, because it's not about the food. You can ask about the ingredients when you are at the table.

So people with allergies shouldn't tell the people who will be preparing food for them about their allergies before the food is served?

Exactly. Because being invited for dinner is not about stuffing your belly for free, it is a social meeting with more important matters. So if you have an allergy or a diet, you can mention that when you're seated. The host can make something that suits you, and if they aren't able, you can just have the drinks. That's why you eat before going out to dinner, so you don't have to worry about your belly. And it is also very convenient in case if the cooking is just bad.

I've noticed that these kind of social rules and politeness has been increasingly lost in people, and it is because of widespread narcissism. It's "me, me, me". The result is that people don't invite each other to dinners or other social gatherings, and everybody is worse off because of that.

Re: Ask HN: What non-AI products are you working on?

#424
I‘m learning ImGui currently and try to build a data visualization tool around SQLite that works cross platform. Even webassembly (using emscripten) would be great, but I can‘t manage to establish cross platform networking right now. For example to import JSON data from a remote url.

Re: Ask HN: What non-AI products are you working on?

#426
I've started tinkering on a tool for test case management. It would be a direct competitor of things link TestRail or Practitest. QA folks that I talk to seem to feel very underserved by these types of tools, so it feels like there would be demand for something with a bit more polish.

Check it out: https://qually.app/

Re: Ask HN: What non-AI products are you working on?

#427

A new terminal emulator [0] from scratch. It's native and cross-platform for Windows, Mac, and Linux. This is a product with a history of overpromising on the release date, but I'm being more realistic with the roadmap and streaming progress on Twitch. If it helps think of it as the indie, non-AI version of Warp [1]. [0] https://terminal.click [1] https://warp.dev

Great landing page for warp, also the tool looks really cool (I'm on Windows, so I have to wait).

How is the privacy aspect? Does it use ChatGPT? Do you support any local LLMs?

Re: Ask HN: What non-AI products are you working on?

#428
Just launched "Postcard". My co-founders and I left Google & Meta to radically enhance the way a user connects and interacts with their closest friends & family. Many of the legacy consumer apps us an approach from 10, 15 ,20 years ago and users are looking for higher quality experiences and products. Lots of takeaways after launching our MVP & V1.. and we just submitted V2 to Apple & Google. www.thepostcardapp.com

Re: Ask HN: What non-AI products are you working on?

#429
post #155

Earlier quoted context omitted.

I know it's crass, but why? Aren't terminals good enough

I wish somebody built an emulator as fast as xterm and as configurable as something like kitty. Until then, xterm it is.

xterm has extremely poor throughput performance. kitty and most other well designed terminals are atleast 2x as fast as xterm. For example: https://sw.kovidgoyal.net/kitty/performance/#throughput

Of all tested terminals xterm is faster only than konsole.

Re: Ask HN: What non-AI products are you working on?

#430
post #316
post #103

Long-term (decades), no-subscription archival storage. Essentially, you buy a block of space, upload your data over time, and it gets distributed when you need it (if you lost a primary backup), or on your death (to friends and relatives, or whoever you choose), or on a specific future date. It's a mashup of a safety deposit box, time capsule, and deadman switch. It's not ready yet, but will be ready in a few weeks.…

I worked on this in 2018/19. It's a really hard problem, and the only people I've found doing it at scale is a small Norwegian company called https://piql.no . They encode large files into film and store it in the Arctic World Archive - and they did GitHub's Arctic Code Vault. DNA storage is interesting, though can be damaged by radiation. Would love to see where you land with this project - it's one of the few cases…

I think boring, standard solutions are actually pretty effective. We use commercial cloud cold storage for primary backup, with a self-hosted cache layer in front to drive down cost. Commercial off-the-shelf storage is also the best characterized and has a proven track record.

NIST has some great reviews of the stability of optical media, and it’s quite good, done for the library of Congress.

DNA storage would enable some pretty crazy storage density, but ensuring there’s a compatible reader around in 30 years might be difficult

Post reply on HN