Josh Ramirez

building @ for macOS
interested in computer architecture & systems
Cornell Engineering ’30

[Josh Ramirez @jram4dev]

just released the first public beta of Wield, a free macOS launcher i've been building.

it's built around leader keys. looking for early power users who appreciate speed without waiting around for ui.

free, all on device. eager for feedback.

https://t.co/Vs0v8rVmrc

[Josh Ramirez @jram4dev]

@janusch_patas must have formed an rl loop of 8 year old video game lovers as the reward function

[Josh Ramirez @jram4dev]

@Spotify just re-bought the spotify premium subscription. listening to weird fishes on youtube music just wasn't enoupgh. spotify has the special sauce that others miss. I miss you so much, even looking forward to each day usnig spotify all I want to do is be with you.

[Josh Ramirez @jram4dev]

@migtissera gpt with cyber might be on to something further than the frontier anthropic. they just need to release it to satisfy all the users that have been wating for weeks and weeksssss

[Josh Ramirez @jram4dev]

@inferredbylisa how much time and effort really went into your startup. were you fundamental or more of an overseer. this directionally feels pretty bad, especially from friends or even best friends

[Josh Ramirez @jram4dev]

@tim_cook the magic from moving from windows to a macbook has been unmatched. even building further on top with launchers or https://t.co/Vs0v8rVmrc have just pushed it further. never going back to a windows laptop or even computer

[Josh Ramirez @jram4dev]

@robj3d3 and it will just keeping coming. can't wait for anthropic to take the monopoly on regeneron and pfizer too. then come for collges which will have integrated by that point

[Josh Ramirez @jram4dev]

@Dan_Jeffries1 i loooooooveeeeeeeeeeeee taking down the american economy. its possible the blackwell leaks are a true reason for the jensen shakedown

[Josh Ramirez @jram4dev]

@Conor_D_Dart can't loose respect for @thsottiaux. we thought @OfficialLoganK and gemini had fallen so far behind yet today hyrdrogen isint looking too bad. the frontier labs are never lost, just depends on the cycle

[Josh Ramirez @jram4dev]

@thdxr we need neuralese, the agents long for integrated fuulllly rl-ified language

[Josh Ramirez @jram4dev]

@dummerspast39 its not all about straightness, tibo had a questionable period along with darios questionable role in nvidia news right now. just spreading rumors...

[Josh Ramirez @jram4dev]

@Coinvo and no more datacenters along with massive power restrictions because of the GWP of datacenters not factoring in the solution to global warming

[Josh Ramirez @jram4dev]

@thsottiaux better subagent handling, what i've learned form t3 code is the appreciation of unified steering and the agents.md seems to listen better. this may be an artifact of the codex prompt. but auto steering is also nice. whenever i type a prompt when running 90% of the time i want steer always

[Lex Fridman @lexfridman]

NVIDIA letting Jensen go is a huge mistake, for America and for humanity. It's Apple firing Steve Jobs all over again, but much bigger, given the role of AI in defining the future of our world.

I've been keeping quiet on this out of respect for Jensen's family and team, who I've met and who are amazing humans, and because I personally hate infidelity. So my heart goes out to them in this difficult personal family moment.

But I can also say that Jensen is human, and makes mistakes like all of us do. Not that I matter at all, but I've known Jensen for a long time and have gotten the chance to spend a lot of time with him, behind the scenes, away from mics and cameras. I have seen him show kindness, humility, care to everyone around him, including strangers, including from the poorest most struggling parts of society, treating all the same with the same thoughtfulness and attention, because they are fellow humans and not because he has something to gain from them. That shows rare character, to maintain humility and humanity through all of it.

Another data point is what people who know him well have said privately to me about Jensen over the years. I've never had anyone tell me anything but how awesome he is or how much they love him. Probably over a thousand people at this point. This includes hundreds of engineers, CEOs, colleagues, friends, etc.

Also, since there was some spy talk. Jensen is as proud an American as it gets. I'm in shock frankly at all of this, including the mass hysteria around it. I love America, and I think NVIDIA letting Jensen go is terrible for America.

I stand with Jensen.

Love you all ❤️

[Josh Ramirez @jram4dev]

@semichenkko makes you wonder just how much might be possible to unlock with fable 5.2 or sota with the right prompting and harness

[Josh Ramirez @jram4dev]

@tszzl this is really bad is china further behind. is there anyway to recover from this. brick the chips from afar. maybe state kill switches in graphics card do serve a purpose

[Josh Ramirez @jram4dev]

@willdepue insane work just for the love of the game. big if true. deserves many many follows if true. is this already priced in. is it time to short or whats the outcome of this info being public or outcome to nvidia?

[will depue @willdepue]

I have a major result to share.

I’ve found conclusive forensic evidence that the top 4 Chinese AI labs—DeepSeek, Moonshot, MiniMax and Ant Group—have built their newest models on NVIDIA’s most advanced Blackwell chips, which are barred from export to China.

This sounds insane, but I discovered that almost every AI model carries a hidden signature of the GPU that trained it. When a model is first set up, the starting values of its weights are laid out in a pattern that depends on the exact GPU. Each combination of GPU and seed number produces a unique pattern—a sort of fingerprint of the hardware used to create the model. Traces of that pattern survive training into the published model.

Again, this evidence isn’t based on any insider information or leaking. It’s based on a forensic analysis of the model weights released by the companies themselves.

I searched more than six trillion combinations of chip layout and seed number and matched them to open-weights models. In doing so, I found conclusive evidence that six Chinese AI models—including DeepSeek-V4-Pro, Kimi K3, and MiniMax-M3—were initialized on NVIDIA Blackwell systems. Four of the models match Grace Blackwell, the GB200 and GB300 parts in NVIDIA’s flagship liquid-cooled rack-scale systems.

I then verified these results on real hardware. Running unmodified PyTorch with the seeds I’d recovered, a Blackwell B300 reproduced DeepSeek-V4-Flash’s untouched starting weights bit for bit, and a rented GB200 reproduced Kimi K3’s to a correlation of 0.999995. A Hopper H100 reproduced none of DeepSeek-V4-Flash’s, and changing the seed by one broke the match. (DeepSeek’s seed is 953049583 by the way.) The method also was verified to match 24 known open-weights models to the hardware that developers said they used.

It is exceedingly unlikely that these results are a coincidence. Across six trillion combinations, the best accidental match should reach only about 8 sigma (i.e., eight standard deviations above what random chance would produce). The real matches range from 49 to 1,385 sigma, and even the weakest single test hits 20.

These were data-center-scale training runs. With some analysis of release schedules and model details, I estimate that the largest, Kimi K3 and DeepSeek-V4-Pro, used 2,000 to 3,000 Blackwell GPUs each. Clusters of this size cost on the order of $150 to $200 million each to buy, and across these labs alone would add up to more than half a billion dollars of export-controlled GPUs, not counting experimental compute or otherwise. It’s quite likely that we see multi-billion dollar expenditure on export-controlled GPUs.

I am highly confident in this result, and it has been independently verified by senior researchers at OpenAI and Anthropic. You can read the full report at https://t.co/YOtzND0Jqk.

There is much more I’ve identified about these models through other forms of checkpoint forensics like this, and will have more to share soon.

[Josh Ramirez @jram4dev]

@OfficialLoganK @GeminiApp @GoogleAIStudio @antigravity gemini is so back! happy to see it. i still remember the old days of building with gemini 2.0 flash in the api. the only ones to have a true generous free tier for api builders

[Josh Ramirez @jram4dev]

@GoogleDeepMind @ArtificialAnlys gemini hydrogen carbon argon potassium calcium, elemental rsi?

[Josh Ramirez @jram4dev]

@VictorTaelin likely much more efficient to manage a universal port improvement loop or prompt that you can run every time with a new model