17y in engineering / Indie Hacker / Car mechanic updatify.io jellybox.app

Earth
Today I turned 35 ๐ŸŽ‰๐ŸŽ‚ A bit depressing number honestly, feels like half of my life is done But I hope next half will be much better than previous
308
9
729
60,504
Slowly growing From 250 to 450 in few days
1
4
141
Didnt even know until someone told me my OSS github.com/avdept/JellyBoxPlโ€ฆ got #4 of the day on github, not #1 but still good IMO
2
6
184
@denicmarko you'll like this bro, knowing your respect to OSS apps
1
1
23
UI gets improvements Now you can not just switch playback target between devices, but also control playback and volume, change queue, etc. All syns Also the switch time reduced from ~1.5sec for FLAC files down to 500ms on wifi <-> lte connection and mp3 staying around 250-300ms
79
This is how test lab for JellyBox looks like Unfortunately it's impossible to tear music playback with unit tests
1
1
123
i work at big company(relatively) we do ship features quicker, but we also need to spend more time on maintenance now feature adoption declined - users dont need new features everyday, they just want product to do what it was advertised to do more tech debt to address now, involving reworking whole parts of app(that was heavily rewritted by AI already) tons of PR to review that nobody wants to touch because its hard to understand if it will break something or not(yeah, tests there, but that not always helps) less people understand the product now - one day you open and see new features that you have no idea when added and how they work not complaining, just saying my own feelings
Almost every friend, i talked to, who works as as dev in enterprise or big company, says (along those lines): "With AI we're tearing through backlog like crazy! We ship features that were backburner tasks for years!" Yet, those companies are reporting marginal ROIs on AI integration. Makes me wonder if shipping more features โ‰  better bottom line? Not just at small SaaS level. Because you'd think they'd make more money with more features?
1
7
722
So, VW complains about declining sales and yet they release this bentley which looks more like some chinesse(no offense) RR copy with tons of led strips and useless features rather than just blindly building EV on Bentayga platform
71
Something weird I setup subdomain and app. Subdomain managed by cloudflare, app runs on digital ocean About 1h later I started to get random requests from bots. How come my other apps dont have bots, while app, which domain managed through cloudflare - gets spammed?
2
84
bro never heard about microsoft copilot(s) ๐Ÿ˜‚
Google now makes 3 completely different laptop product lines: - Google Pixelbook - Google Chromebook - Google Googlebook I never understand why Google's branding and naming is always so confusing I'd name it maybe Chromebook Pro?
113
My OSS player Jellybox soon to be better than spotify Got continuity mode working - when you can move playback from device to device in 1 button click
3
186
I hate when you run some OSS app for years and barely have any visibility and then some guy vibe codes slop app and gets 3k starts overnight for barely working app
1
82
My first WAF rule ever these guys were running like 2k requests a day last month, not sure what were they looking for
2
3
196
Would you support me launching my OSS free music player on product hunt? working on it since 2023
2
1
5
247
Not soemthing ppl say you everyday
8
213
Added a cool list update animation The list isnt just updated, but new item insertion has animations
3
175
I need opinions One guy on team produces a ton of PRs with fixes for issues which never happened, edge cases like "IE5 user visits a site and does not see something" These fixes adds a ton of code, conditions, etc I'm curious should these things be merged? On one side its things that might help in future On the other - its the code to maintain and overall performance penalty(not for 1 PR, but after 1000s) Merge or reject?
8
5
643
This is my quickest growth product and yes, its opensource and free
2
8
367
One of the downsides of "AI lobotomization" - is that our new knowledge becomes shallow means we understand some high level concepts without understanding why and how and thats bad because without understanding topic - we can't efficiently prompt it, because as you know LLMs do the job best when its prompted with enough details, not generic "Make it good and make no mistakes" Recently I understood that while working on custom game engine built from scratch. One of issues I had with shadow rendering and near culling. I learned these concepts later when I wasted 2 days trying to make LLM to fix it by explaining in simple words - like "There's area without a shadow near camera" LLM tried tons of ways, to fix it until I actually read an article about shadows, culling, camera far/near planes in game engines. From there I knew whats the issue and solution took literally few minutes. A very simple example but it kinda showed me the deeper problem with offloading deep learning process to LLMs
not a popular thought, but AI "lobotomizing" us simply because you dont need to make brain to work and as every muscle - brain gets atrophied over time I think we'll see consequences in 3-5 years
2
5
302
not a popular thought, but AI "lobotomizing" us simply because you dont need to make brain to work and as every muscle - brain gets atrophied over time I think we'll see consequences in 3-5 years
We thought AI would make us more productive But what if it was just making us lazier? We do more in less time, but the average person will probably just... work less
4
1
7
749
Giving some proper test to android auto companion app for jellybox
2
115