Sukhavati

5.8K posts

Sukhavati banner
Sukhavati

Sukhavati

@avatisukh

God made the desert to test the faithful

Katılım Ocak 2012
1.5K Takip Edilen432 Takipçiler
A-100 gecs
A-100 gecs@PinstripeBungle·
@avatisukh hey I like to think I was a high IQ meat puppet with no agency, thank you
English
1
0
171
6.4K
A-100 gecs
A-100 gecs@PinstripeBungle·
one thing about the modern biometric passport system and centralized immigration software is it means these small mercies done out in the field by one person's principled stand aren't possible anymore. you literally cannot be the modern Raoul Wallenberg for technical reasons.
Our Passports@ourpussports

1940 life-saving transit visa issued to a destitute Jewish family going to the US. The visa was issued at the Japanese embassy in Sofia, Bulgaria, with the long journey going east via the USSR.

English
26
617
6.2K
195.3K
Sukhavati
Sukhavati@avatisukh·
@Read_Acted why is everyone using US costs of transplant in Spain when there’s no international efficient organ market and spanish doctor salaries are a fraction of US salaries? directionally correct but quite easily dismissed imo, we can do better
English
0
0
1
12
🐍 Ƥşץ𝕔𝔥𝕠pomp_Đ𝔦o̷͝ηуรǗ丂 ඏ
8 Million Euros for the organs of one euthanized r@pe victim in Spain That’s about 25x the lifetime tax contribution of the average Spaniard R@pe is inefficient though. More sustainable to just harvest invaders instead, as there is a near endless supply.
🐍 Ƥşץ𝕔𝔥𝕠pomp_Đ𝔦o̷͝ηуรǗ丂 ඏ tweet media🐍 Ƥşץ𝕔𝔥𝕠pomp_Đ𝔦o̷͝ηуรǗ丂 ඏ tweet media
🐍 Ƥşץ𝕔𝔥𝕠pomp_Đ𝔦o̷͝ηуรǗ丂 ඏ@Read_Acted

So the euthanasia thing is actually an industry. Like transgenderism, like circumcision. A devouring monster. In this case, organ harvesting and getting a liability off state welfare (Dickensian!) It’s so bold I’d support it if I could ensure only my enemies participated.

English
1
7
86
1.6K
Sukhavati
Sukhavati@avatisukh·
@vladimirxyimir @FedeItaliano76 oh i’m sorry hadn’t realized i’m GETTING MONEY because i’m a GRIFTER, where’s my check ELON???? no one knows what love for the game is anymore
English
0
0
0
43
Federico Italiano
Federico Italiano@FedeItaliano76·
Artwork by Anatoly Fomenko, a Soviet and Russian mathematician born in 1945, who is also recognized as a topologist and the creator of the pseudoscientific theory known as 'New Chronology'
Federico Italiano tweet media
English
45
923
8.6K
339.9K
Sukhavati
Sukhavati@avatisukh·
@CharlotteFang77 yeah this was an absolutely retarded claim. as if philosophy became obsolescent after 60 years was such an epic pwn. remilia is rooted in nurturing praxis upon theories that have regrettably wilted under the scorching desert of ideology
English
0
0
3
379
♡ Charlotte Fang 🪲 Crown Prince ❀ LOVE HEALS 💞
Did you seriously publish your claude conversation trying to critique Remilia, and the only argument it could come up with is we build on a wide body of philosophy that we’re fully open about and frequently discuss (where do you think it picked up the references?)— Chardin, Barthes, Debord, Guenon, Land, Fisher, Evola— let Claude know they also forgot Stewart Brand, Ted Nelson, Timothy May, William Burroughs, Nietzsche, Hayek, McLuhan, Deleuze. Not to mention Augustine, Confucius, Tocqueville, Mill, Carlyle. You should try reading them, you might learn something.
banteg@banteg

remilia is 60 years behind the entirety of the milady collective philosophy is just repackaged ideas from sixty years ago. read more: remilia-60.pages.dev

English
44
30
567
28.3K
Sukhavati
Sukhavati@avatisukh·
@an0n_Nic @allenf32 blud I agree completely but the FT comments have some of the most wildly retarded takes available online
English
0
0
10
1.2K
Sukhavati
Sukhavati@avatisukh·
@CigsMake if a couple dozen million people tried to acquire that kind of desk it would permanently collapse various ecosystems and cost tens of thousands of dollars. hence we have ikea desks
English
1
0
2
91
Cigarette Nostalgia
Cigarette Nostalgia@CigsMake·
Desks from the 1800s are still standing while your IKEA desk has disintegrated after a 2.5 years
Cigarette Nostalgia tweet media
English
203
1.5K
18K
369.6K
Veetsam
Veetsam@veetsam·
@LeoNelissen I am not very familiar with dutch politics,but as far as i know the decision to close the Groningen Gasfield was not so much about Green House Emissions,but more about the negative seismic effects caused by extraction.I would not call it a dumb decision,but a complex issue.
English
25
2
103
25K
Sukhavati
Sukhavati@avatisukh·
@spandrell4 no free lunch sadly as these offgas quite a bit, comparable to other couches and mattresses but nonetheless
English
0
0
4
732
ak0
ak0@annanay·
On Agentic Trading Over the past few weeks, I've been inundated with DMs asking if I am seeing anyone have any success with agentic trading. Here were my initial thoughts: >Infrastructure is not a problem. AI is good enough to one-shot an event-driven trading strategy with order management system now. That's actually how we built our testing framework at @QFEX. >Event predictions don't work. Think carefully about what it means to trade on PMs with AI. PMs move on touchdowns [sports], botched speeches [elections], and geopolitical news [war]. How is an AI agent going to predict that? >Agents are okay at backtesting and iteration? Most trading strategy iteration is hypothesis --> model --> test. CTA funds have generated billions in returns from fairly simple ideas following the above playbook. This is because the markets they trade - commodities and indices mostly - have slow regime changes and supply/demand shocks that absorb over weeks over deep liquidity, not milliseconds over $100 of event contracts. The data to trade these strategies is already available via providers like @DatabentoHQ , who we use as an orcale. Another reason to be bullish is that the neural nets a lot of long-term quant hedge funds they use don't look that different to the underlying architecture of most LLMs. Many ways to skin a cat. This doesn't mean HFT strategies are out of the picture, but mid-frequency daily/ weekly type stuff that doesn't rely on nanosecond-precision timestamps and L3 data certainly has a better chance of producing alpha right now. I think we'll see the first major AI-native CTA fund success this year. To help them along, we abstracted away the infrastructure part and made an agent-friendly CLI for people to trade directly on @QFEX, 24/7/365. Happy trading.
QFEX@QFEX

Agentic trading live today through our CLI. Works out of the box with @claudeai or Codex by @OpenAI.

English
8
5
90
15.7K
Sukhavati
Sukhavati@avatisukh·
@KorwinD_El @KnowTopologies @FedeItaliano76 that’s the theory he builds on top of dozen intriguing facts that point to at least some elements in history being counterfeit or wildly misdated. you can build a much less radical theory on top of these facts
English
0
0
1
36
Corwin El
Corwin El@KorwinD_El·
@KnowTopologies @avatisukh @FedeItaliano76 Total bullshit, in general his ideas can be described loosely as "actually, we live in ~700 CE and everything else in official chronology is just a bunch of overlapped events and historical figures". Like Alexander Macedonian and Jesus are the same person.
English
1
0
3
153
Sukhavati
Sukhavati@avatisukh·
@banteg does uv prevent this? I have seen claims it does but I can’t track down a good explanation
English
3
0
0
1.1K
banteg
banteg@banteg·
litellm exploit was using an extremely legacy python packaging feature that predates venv and pip. i never heard of it, yet this footgun is still out there. it runs code on interpreter starts, the "precaution" it uses is the line needs to start from "import", but nothing prevents you from writing "import this; my_malicious_code()" .pth support in python must be killed immediately to prevent further harm.
banteg tweet media
English
10
15
227
26.6K
Sukhavati
Sukhavati@avatisukh·
@RokoMijic @FedeItaliano76 It can be disproved, and the astronomical evidence supports at least a couple dozen key moments in history being quite off, most notably the Peloponnesian War and the solar eclipse witnessed are wildly off target
English
0
0
7
424
Sukhavati
Sukhavati@avatisukh·
@swyx @karpathy how does uv help prevent this? not aware of their initiatives for safety
English
0
0
0
413
swyx
swyx@swyx·
@karpathy we should probably also treat this as a wake up moment for all noveau package managers - uv and bun presumptively - to make these entire classes of things far less risky, eg by adding a lot of guards on install scripts up to the point of manually approving baches of network calls
English
16
14
510
141.6K
Andrej Karpathy
Andrej Karpathy@karpathy·
Software horror: litellm PyPI supply chain attack. Simple `pip install litellm` was enough to exfiltrate SSH keys, AWS/GCP/Azure creds, Kubernetes configs, git credentials, env vars (all your API keys), shell history, crypto wallets, SSL private keys, CI/CD secrets, database passwords. LiteLLM itself has 97 million downloads per month which is already terrible, but much worse, the contagion spreads to any project that depends on litellm. For example, if you did `pip install dspy` (which depended on litellm>=1.64.0), you'd also be pwnd. Same for any other large project that depended on litellm. Afaict the poisoned version was up for only less than ~1 hour. The attack had a bug which led to its discovery - Callum McMahon was using an MCP plugin inside Cursor that pulled in litellm as a transitive dependency. When litellm 1.82.8 installed, their machine ran out of RAM and crashed. So if the attacker didn't vibe code this attack it could have been undetected for many days or weeks. Supply chain attacks like this are basically the scariest thing imaginable in modern software. Every time you install any depedency you could be pulling in a poisoned package anywhere deep inside its entire depedency tree. This is especially risky with large projects that might have lots and lots of dependencies. The credentials that do get stolen in each attack can then be used to take over more accounts and compromise more packages. Classical software engineering would have you believe that dependencies are good (we're building pyramids from bricks), but imo this has to be re-evaluated, and it's why I've been so growingly averse to them, preferring to use LLMs to "yoink" functionality when it's simple enough and possible.
Daniel Hnyk@hnykda

LiteLLM HAS BEEN COMPROMISED, DO NOT UPDATE. We just discovered that LiteLLM pypi release 1.82.8. It has been compromised, it contains litellm_init.pth with base64 encoded instructions to send all the credentials it can find to remote server + self-replicate. link below

English
1.3K
5.4K
27.9K
65.6M
Sukhavati
Sukhavati@avatisukh·
@Crypto_McKenna since vague post king won’t post source here’s what I think he’s referring to, analysis of 1952 plates from Harvard observatory before there were any satellites. transits observed in the Earths shadow are assumed artificial satellites of unknown origin scholar.google.com/citations?view…
English
0
0
12
975
McKenna
McKenna@Crypto_McKenna·
Can’t tell if this is exciting or extremely alarming
McKenna tweet media
English
18
2
71
27.2K
ps
ps@psus24·
@SAshworthHayes Given the road fatalities per 100k driven over here. I have to say 57 lives a year through car seats is utter rubbish We have an accident rate that bends European minds (I’m European in U)
English
6
0
36
9K