HNHacker News
TopNewBestAskShowJobs

stevenhubertron

1,005 karma · joined June 23, 2014

submissionscomments
stevenhubertron··on Why Doesn't Anyone Want to Fix One of America's Scariest Roads?
What an unhelpful comment.
stevenhubertron··on macOS Golden Gate Is a Buggy Mess
It’s been an improvement for me.
stevenhubertron··on Qwen 3.8 Omni Flash
Pick the cheapest model with good speed. ZDR, and price. If it works great. If it doesn’t pick one a bit more expensive till you get what you need.
stevenhubertron··on iOS 27, iPadOS 27, and macOS 27
Not entirely but its ALOT better then it was.
stevenhubertron··on MacFileSizing – What is consuming my disk space?
Why is their better than Mole or Grand Perspective?
stevenhubertron··on Claude, change the "Add to Cart" button to blue
You can hate AI all you want, but this is because of a bad design system, not because of a bad LLM.
stevenhubertron··on I-have-ADHD: A skill to stop coding agents from burying the answer
Doesn’t work
stevenhubertron··on An update on leaving Gmail for Fastmail
I have used Fastmail for 5+ years, no real complaints other then some real emails get caught up in spam folder.
stevenhubertron··on AI;DR (AI; Didn't Read)
Seem's reasonable and what I hold myself and my team to.
stevenhubertron··on Claude: System Prompts
Man that is alot to be sent every turn.
stevenhubertron··on Show HN: AI Pulse a fake LED strip beside the macOS Dock that shows agent status
Edit: put in a PR for PI support: https://github.com/leog/ai-pulse/pull/1
stevenhubertron··on Show HN: AI Pulse a fake LED strip beside the macOS Dock that shows agent status
Looks like this: https://www.drkpxl.com/assets/img/ai-pulse.png in the lower right of the screen.
stevenhubertron··on Show HN: AI Pulse a fake LED strip beside the macOS Dock that shows agent status
Shame its for Claude only and not a more universal harness. Also wish their was an actual screenshot of it in action.
stevenhubertron··on Muse Glimmer: 30B-parameter model optimized for always-on local agent workflows
For so many non-coding workflows, tool calling is more important.
stevenhubertron··on Show HN: Qwen Scribe – local transcription and dictation for Apple Silicon
Lotsa alternatives out there that are fully Mac native like https://voice.baselinemakes.com
stevenhubertron··on Show HN: Yap – OSS on-device voice dictation for macOS with no model to download
This does the same thing with Apple models or Ollama models. Also has meeting transcription: https://voice.baselinemakes.com
stevenhubertron··on Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor
It’s a great API. I use it in my app and it’s feels so much more accurate.
stevenhubertron··on Building and shipping Mac and iOS apps without opening Xcode
This is how i am developing keptrecipes.com. Been a pleasure this way TBH.
stevenhubertron··on Dictation and Meeting App Sends to Apple Notes – Free and Local Models
Repo here: https://github.com/drkpxl/better-voice I cleaned it up a bunch as I did like 30 iterations of this before I was happy with it.
stevenhubertron··on Dictation and Meeting App Sends to Apple Notes – Free and Local Models
Hi I built this things for myself because I didn't want to have both a dictation app and a meeting app I had to pay for. After sharing it with some friends they said I should put it out there since it's free and solves problem lotsa paid apps do. There are certainly some rough edges but would love your thoughts.
stevenhubertron··on I used to love Claude, but the latest models are slowly ruining it
Same, I was just fighting it as it accused me over and over again of Ctrl+Cing a process that clearly errored out. 3 turns for it to find why the shell script actually crashed.
stevenhubertron··on Most slopcode projects are abandoned and deleted within months of release
Wait till you hear about about CMS migrations I did before AI existed.
stevenhubertron··on Long Island's decommissioned nuclear power plant
I grew up there. I was maybe 14 so I have some memory of how worked up the community was. I remember people talking about building a bridge to CT since there would be no other way to get people off the island. It was such a fierce time then, nothing compared to nowadays about seemingly anything though.
stevenhubertron··on Qwen 3.6 27B is the sweet spot for local development
Can you say a bit more about this? The bad tool calling has made me give up on using Gemma for my Hermes and a personal recipe site. I have only downloaded from Ollama.
stevenhubertron··on GLM 5.2 vs. Opus
No one has really talked about hybrid and using Opus to plan and orchestrate GLMs work both through initial build and code reviews. That’s a true best of both worlds and there doesn’t need to be a winner.
stevenhubertron··on MiniMax M3 vs. GLM 5.2: Codegen comparison across autonomous coding tasks
There are some out there I couldn’t find one quickly, but deep seek reasons way too much
stevenhubertron··on MiniMax M3 vs. GLM 5.2: Codegen comparison across autonomous coding tasks
This has been my experience as well to the letter.
stevenhubertron··on GPT-5.5 hallucinates 3x more than MIT-licensed GLM-5.2
The more I have been using 5.2 the more I have been impressed with it. And I’ve just been using the usually neutered ollama version.
stevenhubertron··on What the Fuck Happened to Nerds
I think it started with “developers developers developers” and kids went to colleges for CS degrees to get rich not to solve problems or geek out on hard problems but to get wealthy. That dropped in a lot of new personalities some of which were highly narcissistic.
stevenhubertron··on Company Will Add Phone, AirPod, and Smartwatch Trackers to ALPRs
Seems like a great use for Fable ;)
Page 1 of 9Next →