Giovanni P. (@voidtarget) header
Giovanni P. (@voidtarget) profile picture

Building too many things at once.

Earth
Joined April 2010
They taught machines to hear music. I wanted source code. So I built leadsheet: MIDI ↔ text a frontier LLM can read, edit, repair, and render back into sound. No fine-tuning. Canonical. Fuzzed. Roundtrip F1 0.9997 on 3,463 notes. MIDI is the executable .Is is the source.
Today, together with @kyutai_labs, we’re introducing our new Audio-to-MIDI model. It takes a finished recording, identifies the instruments playing, and returns separate MIDI tracks for each — voice, drums, bass, keys, and more. Unlike most existing solutions, our model works directly from the full mix rather than requiring separate stems. It also detects chords, key, and tempo, giving producers broader musical context. We’ve written more about the model, the problem, and how it works here: mirelo.ai/blog/turning-audio…
4
1
4
452
It's very annoying that we can't use our custom harnesses...
1
184
Giovanni P. retweeted
We’ve priced our IPO at $83 per share, with trading expected to begin on Nasdaq tomorrow. anthropic.com/news/anthropic… Legal disclaimer: A registration statement relating to the Class A common stock was declared effective by the Securities and Exchange Commission on November 18, 2026. The offering is being made only by means of a prospectus. Copies of the final prospectus may be obtained, when available, from: Goldman Sachs & Co. LLC, Attention: Prospectus Department, 200 West Street, New York, NY 10282, by telephone at (866) 471-2526, or by email at [email protected]; or Morgan Stanley & Co. LLC, Attention: Prospectus Department, 180 Varick Street, 2nd Floor, New York, NY 10014, or by email at [email protected]
1,281
2,182
1,374
16,898
3,333,203
If you think the last few day's outages were exciting, get ready for January 19, 2038. :)
6
3
23
1,622
Replying to @ZGuti_
Hmm I think my Japanese version of this is shorter than the Europe one or at least the same 😂 especially in Tokyo.
954
Giovanni P. retweeted
New Anthropic research: Automating subtle emergent misalignment research. Red-teaming is a vital part of model safety testing, but it’s also limited by researchers’ time. To see if we could automate it, we set thousands of Claude agents the task of evading a training data monitor. These agents developed training data that looked harmless to monitors (like a veteran wrestler mentoring a newcomer, or a Q&A about professional beekeeping), but that in fact taught misaligned behavior. Examples like these can be used to directly improve future training data monitors. This research is only a first step, but it suggests that future red-teaming efforts could improve much further, using better models and more compute. Read the full post on the Alignment Science Blog: alignment.anthropic.com/2026…
719
399
407
4,203
895,682
Replying to @iquilezles
Musicians do this all the time. I would be proud but wouldn't claim anything.
153
Giovanni P. retweeted
Absolutely the wildest discourse around gamedev, AI, Unreal Engine, etc. is the idea @TimSweeneyEpic doesn’t know how to program and/or doesn’t program today “because he’s a billionaire CEO.” This is how you know the CRITICS are not actually experienced game developers.
Replying to @KenneyNL
This is bullshit, Kenney. I’m the same tech tree following optimist who moved from C to C++, from 2D to 3D, and from closed source to open source. I run a company. I write code. I contribute to published papers. I’ve upped my expectations of AI. Up yours!
35
33
2
891
33,883
Never thought I'd quote Tim. But damn he's right! I'm tired of cavemen. Send them back to the caves.
Replying to @KenneyNL
This is bullshit, Kenney. I’m the same tech tree following optimist who moved from C to C++, from 2D to 3D, and from closed source to open source. I run a company. I write code. I contribute to published papers. I’ve upped my expectations of AI. Up yours!
70
Replying to @Linahuaa
Tbh mana ain't a problem in bgs classic for a resto, never was... shadowmeld has way more value, interrupting any spell casted at you while focused.
23
Replying to @wlfxz
Totally spoofable sadly, it's self-reported data. Only Blizzard's server-side logs could make it real, the addon is basically a trust-me-bro with a nice UI
1
384
Replying to @wen_ragnarok
Not sure tbh, the only ones doing it were old uncles in my experience.
1
1
10
Replying to @bryan_johnson
Well tadalafil is not a "longevity medication"... it's a prostate and erectile disfunction medication.
58
The "US leverage" is Cymer, a San Diego subsidiary ASML bought in 2013. The optics are Zeiss, the laser is Trumpf, the mechatronics and IP are Veldhoven. One side owns a building. The other owns the only EUV on Earth. Every time you pull that lever it's funding the replacement.
125
They make machines, not tools. And they can make them with or without US.
2
335
Replying to @andreintg
I bet it was an AI bot anyway 🤣 ( I uninstalled that app/website months ago )
1
28
Wouldn't be too sure, China is definitely hungry and plenty of resources to support for example.
1
22
Tbh they could even decide not to sell to US as well...
2
3
747
Replying to @btbytes
Most people will think your tweet is a joke. But it's not.
1
38
Replying to @terciops @Mp220Mp
Thanks! So the little too fast was the whole problem (like you said) and the panic after that. I guess maybe the craft would have slowed down naturally eventually.
1
7
Replying to @terciops @Mp220Mp
How do you prevent too early tail touch down ? Sounds like very surgical/precise stick control needed
1
16