HNHacker News
TopNewBestAskShowJobs

martythemaniak

11,234 karma · joined February 26, 2007

submissionscomments
martythemaniak··on Gemini 2.5 Flash Image
What is a "1920s nazi officer" what do they look like?
martythemaniak··on Tesla insiders have sold more than 50% of their shares in the last year
It's only about 1-2 percent of most broad market funds, but if you do care it should be relatively easy to hedge. TSLQ is a 2X TSLA short etf, so if you have 100K in index funds, you'd buy like $750 worth of this.
martythemaniak··on Grok Has Been Suspended
Let me guess, it start calling itself Abrodolph Lincoler?
martythemaniak··on Oakland cops gave ICE license plate data; SFPD also illegally shared with feds
> If you're gonna be angry at someone be angry at the people among us were in favor of the creation of this data set

You're right, they're not shifting blame, they're straight up telling you who is to blame. And apparently, it is not the people who came up with this idea, nor the people carrying out the actions, nor the people supporting them.

martythemaniak··on Linda Yaccarino is leaving X
From your list, only grok. All the other stuff was already there.
martythemaniak··on US Court nullifies FTC requirement for click-to-cancel
Yeah, those crazy woke judges that think that the government should not be able to bust into your bedroom and arrest you because you used a condom.
martythemaniak··on Elon Musk's Grok praises Hitler, shares antisemitic tropes in new posts
There's a tweet from Musk a week or two ago that they were specifically training Grok to be this way, it really isn't a failure. I mean, throwing nazi salutes and shutting down foreign aid and cancer research has also been bad for Tesla's business, but nevertheless he persists.
martythemaniak··on Gemini Robotics On-Device brings AI to local robotic devices
OpenVLA is basically a slightly modified, fine-tuned llama2. I found the launch/intro talk by lead author to be quite accessible: https://www.youtube.com/watch?v=-0s0v3q7mBk
martythemaniak··on Gemini Robotics On-Device brings AI to local robotic devices
1. Well, based on Karpathy's talks on Tesla FSD, his solution is to actually make the training set reflect everything you'd see in reality. The tricky part is that if something occurs 0.0000001% IRL and something else occurs 50% of the time, they both need to make 5% of the training corpus. The thing with multimodal LLMs is that lidar/depth input can just be another input that gets encoded along with everything else, so for driving "there's a blob I don't quite recognize" is still a blob you have to drive around.

2. Figure has a dual-model architecture which makes a lot of sense: A 7B model that does higher-level planning and control and a runs at 8Hz, and a tiny 0.08B model that runs at 200Hz and does the minute control outputs. https://www.figure.ai/news/helix

martythemaniak··on Gemini Robotics On-Device brings AI to local robotic devices
You can think of these as essentially multi-modal LLMs, which is to say you can have very small/fast ones (SmolVLA - 0.5B params) that are good at specific tasks, and larger/slower more general ones (OpenVLA - a finetuned llama2 7B). So a rpi could be used for some very specific tasks, but even the more general ones could run on beefy consumer hardware.
martythemaniak··on Gemini Robotics On-Device brings AI to local robotic devices
I've spent the last few months looking into VLAs and I'm convinced that they're gonna be a big deal, ie they very well might be the "chatgpt moment for robotics" that everyone's been anticipating. Multimodal LLMs already have a ton of built-in understanding of images and text, so VLAs are just regular MMLLMs that are fine-tuned to output a specific sequence of instructions that can be fed to a robot.

OpenVLA, which came out last year, is a Llama2 fine tune with extra image encoding that outputs a 7-tuple of integers. The integers are rotation and translation inputs for a robot arm. If you give a vision llama2 a picture of a an apple and a bowl and say "put the apple in the bowl", it already understands apples, bowls, knows the end state should apple in bowl etc. What missing is a series of tuples that will correctly manipulate the arm to do that, and the way they did it is through a large number of short instruction videos.

The neat part is that although everyone is focusing on robot arms manipulating objects at the moment, there's no reason this method can't be applied to any task. Want a smart lawnmower? It already understands "lawn" "mow", "don't destroy toy in path" etc, just needs a finetune on how to corectly operate a lawnmower. Sam Altman made some comments about having self-driving technology recently and I'm certain it's a chat-gpt based VLA. After all, if you give chatgpt a picture of a street, it knows what's a car, pedestrian, etc. It doesn't know how to output the correct turn/go/stop commands, and it does need a great deal of diverse data, but there's no reason why it can't do it. https://www.reddit.com/r/SelfDrivingCars/comments/1le7iq4/sa...

Anyway, super exciting stuff. If I had time, I'd rig a snowblower with a remote control setup, record a bunch of runs and get a VLA to clean my driveway while I sleep.

martythemaniak··on Trump's NASA cuts would destroy decades of science and wipe out its future
> creating an unprecedented crisis

Nope, that's just your brain on right wing propaganda. Nothing of the sort has happened to justify throwing away the rule of law as this administration is doing. It is like murdering someone because you're annoyed that they chew with their mouth open, just an absolutely, wildly disproportionate response to what has happened. Coincidentally (to riff on the GP) it is almost word-for-word the propaganda Orban used in the mid 2010s to consolidate his authoritarian rule of Hungary.

martythemaniak··on Launch HN: Vassar Robotics (YC X25) – $219 robot arm that learns new skills
Neat! Does this work with open source models like pi0 and OpenVLA? How does the inference-time teaching you outline work exactly?
martythemaniak··on PS5 shooter goes from 5 players to bestseller after devs defend game
I don't remember that one, but I loved de_rats (same idea) so much I stopped playing CS when it fell out of favour.
martythemaniak··on My AI skeptic friends are all nuts
At 0 temperature an LLM is a Map<String,String> - a string input (key) will give you the same string output (value) every time. Hypothesis: there exists a key whose value is a complete, working, fully-tested application which meets your requirements 100% and fulfills your business need. This key is the smallest, most complete description of what your application does. It is written in natural language and represents a significantly compressed version of your application code.

My part-time obsession over the last few months has been trying to demonstrate this and come up with a method for finding these magic keys (I even tried to get the LLMs to search for me, lol). What I really want is to give the latest thinking models (200k input, 100k output) a 5-6 page design doc (4k words, 5k tokens) and have them produce a complete 5kloc (50k tokens) microservice, which would show a 10x compression. It's hard, but I haven't seen any reason to think it wouldn't work.

For better or worse, I think this will be close to what IC jobs will be like in few years. Fundamentally, our jobs are to try work with other functions to agree to some system that needs to exist, then we talk to the computers to actually implement this. If we switch kotlin+compiler for design doc+llm, it still going to be somewhat the same, but far more productive. Agents and such are somewhat of a stop-gap measure, you don't want people giving tasks to machines, you want to accurately describe some idea and then let the computers make it work. You can change your description and they can also figure out their own tasks to evolve the implementation.

martythemaniak··on Do Patients Without a Terminal Illness Have the Right to Die?
I would quintiple-check any such claims of abuse. It's usually just clickbait and outright lies.
martythemaniak··on Doge cuts to USAid blamed for 300k deaths – most of them children
> Is there something I don’t know?

How do you know that the people and countries dependent on this aid were also not working on making themselves less dependent? Are you aware that poverty is decreasing world wide and many countries have successfully developed over the last several decades? How do you know that USAID did not accelerate this processes by helping take care of vulnerable people, so the less vulnerable would have more resources to develop faster?

martythemaniak··on [dead]
So is sending random innocent people to Guatemalan gulags and accepting massive bribes. You can't have "illegal" things without rule of law.
martythemaniak··on Tariffs in American History
There's been tons of discussion and several actual laws - IRA, CHIPS etc.

A large part of why you're seeing the Trump insanity is because there's increasingly little connection between reality (ie CHIPS, IRA) and people's perception (ie your comment)

martythemaniak··on Trump's Middle East AI Deal: "Just Count the Server Racks"
> It is possible, given sufficiently strong agreement details (which are not yet public and may not be finalized) and private unvoiced considerations, that this deal contains sufficient safeguards and justifications that, absent ability to fix other American policy failures, this decision is superior to the available alternatives.

This is a great example of how incredibly stupid smart people can be. They take obvious bullshit at face value, then spend their lives arguing over insane made-up details of that bullshit. It'd be funny if they didn't end up enabling the bullshit grifters.

Trump is very simple. Give him praise, give him money and he'll give you some public concession he controls. He operates like every other despot in the history of the world. Nothing new or interesting about it.

You're an Emir and want some chips? Give him praise, give him money, and he'll change the rules so you can have your chips. If your thinking is more complicated than this, it's wrong.

martythemaniak··on OpenAI for Countries
> We’ve heard from many countries asking for help in building out similar AI infrastructure—that they want their own Stargates and similar projects.

Who is this for exactly? The thing about reneging on your agreements and treaties and threatening and demonizing everyone around you is that they learn not to trust you. US-led AI sounds terrible, it would never pass muster in Canada. Neither in the EU, China, India, Brazil... Like, you Cannot entrust your governments functioning on the US anymore, you can just get cut off at any point for no reason.

So who's this for?

martythemaniak··on Gemini 2.5 Pro Preview
Goose by Block (Square/CashApp) is like an open-source Claude Code that works with any remote or local LLM.

https://github.com/block/goose

martythemaniak··on I so hate the phrase "vibe coding"
I like it. It's expectation-free exploration and learning that can yield useful results and be fun.

I had an idea for a little app I wanted to build at work, and while the result seemed like it would be neat and useful, the amount of effort required never seemed worth it, so it sat unbuilt for years. So to test out Cursor, I wrote a ~2 page doc with detailed requirements on what the app should do and how it should do it, then made a mock in lucid chart and gave these two things to Cursor and Claude. It was almost able to single-shot the app, it was up and running in about 30 mins, then I spent a few hours fiddling with the UI and getting just right.

I did not read a single line of code, I have no clue what's in there, but the app works, does what I want it to do, and actually exists unlike before.

martythemaniak··on Port of Los Angeles says shipping volume will plummet 35% next week
Ideally, people posting a video link will include a short summary. HN is heavily text-based and unlike text links, you can't skim/scan and decide whether it's worth to spend 10 mins. It doesn't help that most videos are pretty information-sparse and should properly be a 2-3 min read.

Anyway, I don't mind the downvotes, but it'd be nice if people started including summaries with video links.

martythemaniak··on America's financial system came close to the brink
I'd say it's very much by design because it's a common feature of places America's rulers love - Hungary, Russia etc. One of the many ways in which uppity PMCs get put in their place is by the rulers flaunting their lawlessness and gaslighting the PMCs, while of course making PMCs submit to the rules and work with reality.

So here's we have blatantly illegal behavior by the party that constantly sings about "law and order" and the grating helplessness you feel is what they want you to feel, because they are your masters and want to yo always be aware of this.

martythemaniak··on The Llama 4 herd
LeCun fundamentally doesn't think bigger and better LLMs will lead to anything resembling "AGI", although he thinks they may be some component of AGI. Also, he leads the research division, increasing context length from 2M to 10M is not interesting to him.
martythemaniak··on The Llama 4 herd
I heard reality has a well-known liberal bias.
martythemaniak··on Waltz's team set up at least 20 Signal group chats for crises across the world
No, the government has not had issues running military operations using its existing comms. The actual story is that they used Signal on purpose to bypass required government record-keeping laws.
martythemaniak··on Charlie Javice convicted of defrauding JPMorgan in $175M startup sale
Luckily for her, pardons are on sale now.
martythemaniak··on Kill your Feeds – Stop letting algorithms dictate what you think
Bluesky is pretty much that?
← PreviousPage 7 of 34Next →