joao pires

{astro}physicist, msc comp. science student @ ufmg
@cruzeiro

[Dmitrii Kovanikov @ChShersh]

For my non-tech friends, this is like Ronaldo joining Manchester City

[sui ☄️ @birdabo]

everybody calm down.

i got this.

[vids that go hard @vidsthatgohard]

Marrying dua lipa and he might become james bond, dude is on a generational run

[pilotos com frases aleatórias @F1das_frase]

toto: seloco 32 anos sem o chefe podíamos fazer uma homenagem ne

antonelli: eu sou a homenagem

toto: que

antonelli:

[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.

[William De Lucca @delucca]

ontem karl marx dormiu feliz

[BURKOV @burkov]

LLMs process text from left to right — each token can only look back at what came before it, never forward. This means that when you write a long prompt with context at the beginning and a question at the end, the model answers the question having "seen" the context, but the context tokens were generated without any awareness of what question was coming. This asymmetry is a basic structural property of how these models work.

The paper asks what happens if you just send the prompt twice in a row, so that every part of the input gets a second pass where it can attend to every other part. The answer is that accuracy goes up across seven different benchmarks and seven different models (from the Gemini, ChatGPT, Claude, and DeepSeek series of LLMs), with no increase in the length of the model's output and no meaningful increase in response time — because processing the input is done in parallel by the hardware anyway.

There are no new losses to compute, no finetuning, no clever prompt engineering beyond the repetition itself.

The gap between this technique and doing nothing is sometimes small, sometimes large (one model went from 21% to 97% on a task involving finding a name in a list). If you are thinking about how to get better results from these models without paying for longer outputs or slower responses, that's a fairly concrete and low-effort finding.

Read with AI tutor: https://t.co/MipHHO6rjX

Get the PDF: https://t.co/XQrqiaGwIO

[TDM (e/λ) (L8 vibe coder 💫) @cto_junior]

Defending my Spring Boot Java app that uses 64GB RAM to return { "status": "ok" }

[joao pires @jppirest]

@Pinheiro314n eu usava muito o dbscan e o hdbscan. métodos hierárquicos e com n_min samples funcionava melhor... mas vale ressaltar que era em cima de dados de astrometria/fotometria, não espectrais.

[joao pires @jppirest]

weed: R^3 -> N^3

[joao pires @jppirest]

linguistics ⊆ maths

[julinha @juli32bit]

gnt seguinte esse video acabou c a minha vida hj ent vcs vejam tbm

[Zoe Alannah @zoealannah]

AWS engineers right now

[The Nobel Prize @NobelPrize]

BREAKING NEWS
The Royal Swedish Academy of Sciences has decided to award the 2025 #NobelPrize in Physics to John Clarke, Michel H. Devoret and John M. Martinis “for the discovery of macroscopic quantum mechanical tunnelling and energy quantisation in an electric circuit.”

[Ian Goodfellow @goodfellow_ian]

Moonlake is launching publicly today, focused on generating simulations and games. Simulations are an important application of generative models that I’ve advocated for since 2014

[Jacques.web3 @jacques_web3]

Software engineer career path:

- Intern ($50k)
- Junior ($80k)
- Senior ($150k)
- Staff ($250k)
- Farmer ($30k but actually happy)

[joao pires @jppirest]

@ignWest @statsmeister1 fallen didn't have a lot of money back then, but had family support.
fer, on the other hand, had no money and even had to play (and win) to pay for food and travel. I think this was the same for lucas1 and hen.
fnx is a whole different story, but also not as wealthy as cold.

[joao pires @jppirest]

percebo que no não possuir encontram-se a paz e a calma

[joao pires @jppirest]

I've seen heaven without dying
met the devil without trying
and they both seem to want to talk to me