ɐsɹǝʌǝɔıʌ

137 posts

ɐsɹǝʌǝɔıʌ

ɐsɹǝʌǝɔıʌ

@coffeeshopperr

Staff ML Engineer @ Upwork. Health Informatics Research @ Karolinska Institute, ex-MIT and med student as a hobby. building cool side-projects.

Katılım Ocak 2026
81 Takip Edilen27 Takipçiler
ɐsɹǝʌǝɔıʌ retweetledi
Linus ✦ Ekenstam
Linus ✦ Ekenstam@LinusEkenstam·
NVIDIA just killed the awkward pause in voice AI 😱 PersonaPlex 7B is a real-time conversational model that listens AND speaks simultaneously. Like actually interrupts you mid-sentence like a human. Beat Gemini Live on dialog naturalness. 18x faster interruptions. 100% open source. Run it locally. No API bill. No latency.
English
174
375
3.5K
376K
ɐsɹǝʌǝɔıʌ
ɐsɹǝʌǝɔıʌ@coffeeshopperr·
I'll save the paper to read later but on: > Their answer is the Random Neighbor Score (RNS). The idea is elegant: generate biologically meaningless sequences by randomly shuffling the residues of real proteins—preserving amino acid composition but destroying all evolutionarily meaningful interactions. How did they define "evolutionarily meaningful interactions"? MSA?
English
0
0
0
200
Jorge Bravo Abad
Jorge Bravo Abad@bravo_abad·
When a protein embedding is indistinguishable from noise Protein language models have become the backbone of computational biology. Feed them an amino acid sequence, and they return a dense vector—a compact numerical fingerprint that downstream models use to predict function, structure, localization, or the effect of a mutation. The assumption, largely unquestioned, is that this fingerprint actually encodes meaningful biology. Prabakaran and Bromberg challenge that assumption directly. They ask a deceptively simple question: how do you know whether a given embedding actually represents a protein—or whether it's just noise dressed up as a vector? Their answer is the Random Neighbor Score (RNS). The idea is elegant: generate biologically meaningless sequences by randomly shuffling the residues of real proteins—preserving amino acid composition but destroying all evolutionarily meaningful interactions. Then, for each real protein, measure how many of its nearest neighbors in latent space are these random imposters. A high RNS means the model never learned to place that protein somewhere biologically meaningful. Applied to ESM-2 and ProtT5 across thousands of proteins, RNS correlates strongly with structural prediction quality: proteins with poorly predicted structures have embeddings nearly indistinguishable from random sequences. Downstream tasks follow the same pattern—contact prediction precision drops roughly 40% for high-RNS proteins, and variant effect prediction falls to near chance. Most sobering: between 19% and 46% of the human proteome is underlearned by current models, depending on architecture. Intrinsically disordered regions fare especially poorly across all architectures tested. RNS is model-agnostic and computationally cheap—around two minutes on GPU for 10,000 proteins—making it a practical prescreening step before any embedding-based inference. For R&D teams that routinely use protein embeddings to prioritize variants, annotate novel sequences, or screen large libraries, this has immediate consequences. Running RNS before downstream inference flags proteins where predictions are unreliable, reducing the risk of propagating errors into expensive wet-lab campaigns. It also offers a principled way to identify gaps in model coverage—directly actionable for teams building or fine-tuning their own foundation models. Paper: R. Prabakaran & Yana Bromberg, Nature Methods (2026) — CC BY-NC-ND 4.0 | nature.com/articles/s4159…
Jorge Bravo Abad tweet media
English
4
55
268
17K
Luke Berry
Luke Berry@LukeberryPi·
🚨a biblioteca AXIOS foi comprometida, com 300 milhões de downloads semanais 🚨 se você tá acordando agora no Brasil você tem a chance de salvar o repositório da sua empresa pesquisa no seu lockfile por essas 2 versões: axios@1.14.1 axios@0.30.4 se estiverem presentes, pina uma versão específica abaixo (dependendo da major) e dá merge IMEDIATAMENTE npm install axios@1.14.0 npm install axios@0.30.3 o impacto disso vai ser catastrófico
Feross@feross

🚨 CRITICAL: Active supply chain attack on axios -- one of npm's most depended-on packages. The latest axios@1.14.1 now pulls in plain-crypto-js@4.2.1, a package that did not exist before today. This is a live compromise. This is textbook supply chain installer malware. axios has 100M+ weekly downloads. Every npm install pulling the latest version is potentially compromised right now. Socket AI analysis confirms this is malware. plain-crypto-js is an obfuscated dropper/loader that: • Deobfuscates embedded payloads and operational strings at runtime • Dynamically loads fs, os, and execSync to evade static analysis • Executes decoded shell commands • Stages and copies payload files into OS temp and Windows ProgramData directories • Deletes and renames artifacts post-execution to destroy forensic evidence If you use axios, pin your version immediately and audit your lockfiles. Do not upgrade.

Português
27
170
2.3K
369.2K
ɐsɹǝʌǝɔıʌ retweetledi
Erika Lee
Erika Lee@erikalee·
"I'm at my limit" emotional or claude?
English
336
2.8K
19.3K
464K
ɐsɹǝʌǝɔıʌ
ɐsɹǝʌǝɔıʌ@coffeeshopperr·
@lucas_montano @gvrri boa, faz sentido, vou mandar seu tweet pra umas pessoas, não acho que vai adiantar muita coisa mas custa nada
Português
0
0
1
29
montano
montano@lucas_montano·
@coffeeshopperr @gvrri remuneração é por resultado, bonus do grupo e comissão de conversão vinda através de uma ação idealizada pela pessoa e teremos budget para trafego pago, tu pode impulsionar uma ação se ela trazer resultado (e colher a comissão disso)
Português
1
0
0
149
montano
montano@lucas_montano·
PROGRAMA DE ESTÁGIO pretendo contratar 10 estagiários ao longo de 2026, começando esse mês o programa vai durar 3 meses requisitos: > saber usar claude code > +10k inscritos em alguma rede social > ter uma PJ e conseguir receber em € > estar okay em aparecer no canal benefícios: > um salário mínimo (R$1621) > acesso ao Stupid Button Club > comissão + bônus > Gift Claude Max carga horária obrigatória: ZERO mandar e-mail pro e-mail do perssua
Português
155
50
1.5K
359K
montano
montano@lucas_montano·
tráfego pago não cresce rede social, transferência de autoridade por tráfego orgânico sim quando eu tinha 10k inscritos eu fazia muita collab por conta disso. se alguém com 400k inscritos me pagasse pra me divulgar, e ainda com participação nos resultados, eu iria topar outro ponto é que a carga horária é ZERO
Português
1
0
1
239
ɐsɹǝʌǝɔıʌ
ɐsɹǝʌǝɔıʌ@coffeeshopperr·
pergunta genuína: se a pessoa tem 10k+ inscritos em alguma rede social, não compensa mais pra ela investir no próprio tráfego pago e crescer ao invés de fazer um estágio por um salário mínimo? Se ela tá em começo de carreira e já conseguiu 10k+ audiência, provavelmente algum diferencial ela tem que vale mais de 1600 reais...
Português
1
0
1
458
montano
montano@lucas_montano·
@gvrri é uma interseção interessante né é para fazer acontecer além do código espero que usem claude pra rodarem scripts e automatização de marketing, mas não espero que toquem feature do app
Português
1
0
8
4.3K
Advait Paliwal
Advait Paliwal@advaitpaliwal·
I built Feynman, Claude Code for research. I gave it a question and it came back 30 minutes later with a cited meta analysis. It can also replicate experiments on Runpod, audit claims against code, and simulate peer review. Open source & MIT license, link below
English
135
433
4.9K
319.8K
ɐsɹǝʌǝɔıʌ
ɐsɹǝʌǝɔıʌ@coffeeshopperr·
@TheEconomist when u define happiness as access to public health, social security, education, trust in politicians, etc. these are more like "how does the state of these countries enable happiness" which is not a measurement of happiness itself
English
0
1
1
92