I&#x27;ve been puzzled by something: AI generation can produce an elaborate
 figurine, a cartoon character, even a convincing Super Mario — yet it
 can&#x27;t reliably make a simple wedge ramp so my robot vacuum can climb a
 step.<pre><code> For context: I bought a Bambu P2S but can&#x27;t model. I tried the &quot;describe
 it and get a model&quot; AIs — the output is unusable, you can&#x27;t adjust it,
 it&#x27;s never quite what I meant. I tried having an agent write Python to
 build geometry directly — it tops out at simple primitives.
 
 What finally worked: geometric decomposition. I break a complex part into
 ordered, grouped steps, describe each as a small spec, and let an agent
 execute them in Blender (via blender-mcp). That process turned out to
 abstract into a small engine — the key insight being it converts the 3D
 spatial reasoning LLMs are bad at, into the structured code they&#x27;re good
 at. I wrote it up here: https:&#x2F;&#x2F;github.com&#x2F;zhuchaokn&#x2F;spec-3d-model
 
 My questions:
 - Why is &quot;functional part&quot; generation so much weaker than
 &quot;figurine&#x2F;aesthetic&quot; generation? Is it data (no parametrized-CAD training
 sets), representation (mesh vs B-rep), or evaluation (nobody benchmarks
 &quot;does it print &#x2F; is it watertight&quot;)?
 - Is &quot;turn 3D modeling into code for an LLM&quot; the right framing, or am I
 missing something better?</code></pre>

Why can AI generate Super Mario but not a wedge ramp for my robot vacuum?

Pretty self explanatory. Could you folks shed some light on why these issues keep happening?I’ve noticed most posts and replies are just people coming to their own conclusions based on whatever published data.I think it would be helpful to get some actual, non-corporate&#x2F;marketing information on the goings-on by those that actually see what’s happening on the inside.Thanks

Ask HN: GitHub employees what's going on? Why?

Is it just me or is GPU capacity surprisingly hard to find right now , even in Spot pools? I’ve been trying to get H100s and the Spot pools I’ve checked are pretty much always full.I ended up trying MI350&#x2F;MI355s on DigitalOcean Spot and have been running them for about a week without getting interrupted. Curious if anyone else has tried them? How often are you actually getting reclaimed?Also, where are you guys finding capacity these days?

Where are people finding GPU capacity?

Ask HN: What's the current best research/understanding of healthy weight loss?

Github has been down consistently over the last few months - does it make sense to switch to alternatives?

Ask HN: Alternatives to GitHub

A few hours ago I switched my nameservers to Cloudflare in order to enable R2 bucket serving through my own subdomain, and I found out that it silently had injected a JS analytics snippet in my HTML-only JS-free site textlog.cc — I had to go to the Analytics dashboard, Add the site to the analytics and then disable the snippet. I find this approach entirely invasive, you should opt-in to features like that not have to opt-out. Just a warning out there to folks who might not be aware of this.

Tell HN: Cloudflare silently injects its analytics when you switch nameservers

It was nice to use on vscode with offline Ollama but today notices it is not updated anymore. So, alternatives you use?From github:&quot;Note: The continuedev&#x2F;continue repository is no longer actively maintained and is read-only for all users.&quot;From continue.dev:&quot;Continue has joined Cursor
Continue was acquired by Cursor. Our mission was always to ensure developers are amplified, not automated, and that commitment carries on in the work ahead.It was an honor to build with the Continue community. Thank you to each and every one of you who helped us create a pioneering open-source coding agent.What we built together pushed the boundaries of what AI developer tooling could be, and our open-source codebase remains freely available as a foundation for others.&quot;

Continue coding agent is dead. Alternatives?

Hey everyone, I am a 20 year old college student who was recently diagnosed with seronegative spondyloarthropathy. Over the last 3 years, Ive tried apps, paper journal, Google Sheets, and notes to track&#x2F;catalogue my journey. I always ended up quitting within 2-4 weeks and not keeping accurate track.Is this a common experience or do I just have to be better at tracking? Is there a unique&#x2F;non habit-building way to track&#x2F;catalogue&#x2F;recall?Im building something for this (part of the reason why I am asking) - and if anyone is open to a longer conversation, Ill gladly pay for you time!

Ask HN: I've quit six systems for tracking my illness. What works?

Ask HN: At what point does "clean architecture" become technical debt?

I always was a nerdy, deeply technical programmer. Contributing-to-open-source-and-reading-papers-in-my-spare-time type of programmer.Programming is like a game of chess for me: winning (i.e. delivering a product) is important, but only if I played this game myself. I don&#x27;t enjoy winning if my opponent disconnected. I don&#x27;t enjoy beating people who don&#x27;t know how to move pieces. I don&#x27;t enjoy winning with an engine, and coding agents are basically &quot;winning with a chess engine&quot;.I started to use AI for coding a long time ago, when copilot was first introduced. At first, I was happy: &quot;wow, this can write boring boilerplate and unit tests&quot;. Time went by, agents got better, time spent writing code myself went down. Half a year ago I suddenly realized I hadn&#x27;t written any code for over a month, and I am a full-time software dev without any managing work. Not only did I stop writing code, but agents also became pretty good at proposing plans and architectures, and I was mostly providing missing context and technical details of our infrastructure.One day the realization hit hard. Coding agents removed everything I loved, and - mostly - amplified everything I hated. My role is to gather context from other humans and provide a detailed explanation (similar to a well-written jira ticket) to some super intelligent engineer, then review the result and provide more context&#x2F;explanation if needed. So, basically, a manager.A couple of years ago I could work on a single coding task for weeks. I knew all classes, all functions, I was slowly building up information at a comfortable tempo. Now I am forced to increase my mental throughput multiple times. What took a couple of weeks now takes a couple of hours, and I still need to comprehend that. I always feel tired, I don&#x27;t have enough mental capacity to hold that many abstractions and technical details. So you either autoaccept or get burned out reviewing tens of thousands of lines.One more downside is that code has 0 value. You can code your own OS from scratch, and people will think &quot;ye, I can vibe code that in a weekend&quot;. I don&#x27;t even know what is valuable nowadays.Maybe I am a bad employee. Increasing employer profits was never my goal. I always loved to solve puzzles, and for the last 10 years this was well aligned with my employer&#x27;s desire to increase profits. I never was good as a manager, or brilliant at generating ideas. My power - and passion - was solving complex technical problems with my hands, fast and with good quality. Now it is gone forever. I feel completely lost and unmotivated. I don&#x27;t see any value in learning anything related to computer science, reading papers, contributing to open source. I hate my new &quot;AI delegator&quot; role and I hate my work. I loved coding so much (I use neovim btw) that I was sure I would do this until I die, and now I don&#x27;t know what to do with my life. This is the only thing I know, and, most importantly, the only thing I am&#x2F;was passionate about. Anyone else feeling the same?

Coding Agents killed my identity. How do you feel?

I think Windmill.dev (https:&#x2F;&#x2F;github.com&#x2F;windmill-labs&#x2F;windmill) is really cool, and fully support their desire to be paid for their work.However, their README says:&gt; “Windmill is fully open-sourced (AGPLv3)…”But their LICENSE says:&gt; ... The files under backend&#x2F; are AGPLv3 Licensed, except any snippets of code under the compile flag ‘enterprise’. Those snippets and files are under a proprietary and commercial license. Private and public forks MUST not include the proprietary&#x2F;commercial code mentioned above.I dove deeper into the code, there are around 300 files in the codebase that contain references to enterprise or private features, so it&#x27;s not a clearly separated directory that can just be removed.They have 1.1k forks of their repository that I can see on Github, and I doubt all of them took the time to strip the proprietary code to comply with the license.So at this point, it feels like a licensing trap for anyone that clicks the fork button on Github or is attracted by the AGPL licenseDon&#x27;t get me wrong, I wish more companies made their code open-source if for no other reason other than to be transparent. But calling this mixed source repository &quot;fully open-sourced (AGPLv3)&quot; feels like false advertising that actively leads users to violate the terms unknowingly.Also, even if that line was removed from the readme or fixed, it does not feel right to have: &quot;...Private and public forks MUST not include the proprietary&#x2F;commercial code mentioned above...&quot; and the Github Fork button right next to each other. It&#x27;s like a giant violate-the-license button next to the license.

Windmill.dev is not "fully open-sourced (AGPLv3)"

I noticed that OpenCode harness no longer is including DeepSeek. Does anyone know why?The current free models provided by OpenCode are currently: Nemotron (NVidia), Hy3 (Tencent), Muse Spark (Meta), Ox Alpha (stealth&#x2F;unknown developer), Mimo (Xiaomi), Big Pickle (OpenCode).

Ask HN: OpenCode no longer including DeepSeek?

Seems scary, tyrannical and totalitarian to me.

Ask HN: Why does the US government search through people's phones at the border?

What is broken in robotics software right now?

Robotics Software

Im wondering nearly every day:Is there maybe a reason HN does not have that typical &quot;go to top&quot; or &quot;to top&quot; link at the end of, at least very long, posts?Maybe we can file a change request @dang :-D

Ask HN: Why is HN missing a "go to top" link on long posts?

I&#x27;ve heard from someone who wants to revive an old abandoned project
of mine and asks that I change the license from GPLv3 to LGPL or MIT.
I&#x27;m willing to do that but I took down the original repo from github a
long time ago and there seems to be only a fork that still exists.
What&#x27;s the right way to make it publicly known that I&#x27;m changing the
license?

Ask HN: How to Relicense an Old Project

Should the circular cash flows between the top AI companies, Nvidia, Anthropic, Openai, Google and Meta, be banned because of the systemic risks to the global economy?

Ask HN: What is the evidence for a stock market bubble in AI?

My dad gave me his old T420 thinkpad and as any person who finds an old thinkpad i put ubuntu on it then found that the bios chip has more controller of my laptop then me so i Librebooted it with CH341A and installed OpenBSD btw and now it hosts SearXNG on it and bought the domain opensear dot ch and exposed it to the internet this is probably the most stupidest thing i did so far

SearXNG running on OpenBSD on a ThinkPad t420

Official [0] status [1] hasn&#x27;t been updated as I post this but Downdetector [2] confirms an issue is happening at the same time I was experiencing it.[0] https:&#x2F;&#x2F;status.gitlab.com&#x2F;
[1] https:&#x2F;&#x2F;x.com&#x2F;gitlabstatus
[2] https:&#x2F;&#x2F;downdetector.com&#x2F;status&#x2F;gitlab&#x2F;