Scan ─=≡Σ((( つ◕ヮ◕)つ

Knowledge is power, arm yourself.
Hack the world.
Cross Developer.
My opinions are my own.
Soc català, faig coses
@[email protected]
@[email protected]

[Scan ─=≡Σ((( つ◕ヮ◕)つ @ScanBcn]

"Mientras tú no estabas..." y salgo en la foto del tweet XDDDD #LOOOOOL

[Rubén @enrub_en]

Ok, I was stuck to write the tweet, simply I have no words, it is absolutely crazy how the community is unlocking hardware capabilities.

[Rubén @enrub_en]

Another story about an optimization problem solved by removing the problem instead of optimizing it.

Our transactional email provider gives us 5 days of log history.

Historically, the business asked to see everything. That request ended with us storing every log the provider emitted in our DB and a wildly inefficient view, plus unnecessary load on the system and the database.

Lately people were looking at moving that load to a metrics provider, but wiring them up isn’t simple unless you act as the bridge yourself.

The reasons were legitimate: take the load off the main product. Even so, things were getting complicated for something that conceptually didn’t make sense.

Store every email-provider log in the marketing telemetry service? That would clearly wreck how easy and useful that service is to analyze, drowning it in noise. Maybe it met the technical need, but WHAT WAS THE FUNCTIONAL NEED?

So I called a meeting and sat down with the people who historically used that log view.

Those data were used occasionally to confirm whether a user had been sent the account activation email or not. That was it. All those logs, just to hunt for one very specific kind of error.

Suddenly the needs looked completely different. To cover that, all we need is a simple alerting system, which the transactional email provider already supports, to notify us when there’s an error, or even that specific error.

Not only does the need to store all of that anywhere disappear entirely, we also go from reacting when someone reports the error to being proactive: the alert reaches the right person the moment it happens, and more importantly, long before a user can notice something went wrong.

For the record, I opened the original issue to get those logs out of the main product, for performance and privacy reasons. But I first approached it from a technical angle, not a functional one.

Always ask yourself: why are we doing this?

[Rubén @enrub_en]

Am I the only one who thinks Nike could do its own version of Grok Bot with the motto “Just Do It!” and absolutely crush it, making AI adoption skyrocket?

@elonmusk I’m sure you have a friend there you could share this idea with.

[Rubén @enrub_en]

What is reasoning? 👇

[Rubén @enrub_en]

Using Grok Bot to book a hotel, the first option it gave me was a hotel that was shut down last year for operating without a license, with illegal wiring that caused a fire.

It’s open again now, but man! Isn’t that a very Grok recommendation? Hahaha

[Rubén @enrub_en]

https://t.co/lDnC7QUBvk

[Pere Martí Colom @peremarticolom]

La pluja és la mateixa pels @FGC que per @rodalies. La diferència està en que els primers fa molts anys que estan gestionats per la Generalitat i els segons pel govern espanyol.

[Barrufet del temps @MeteoBarrufet]

No serà un medicà (mini huracà mediterrani), però..., gira com un cicló 🌀 i a vista de satèl·lit ho sembla (a menor escala).

[Scan ─=≡Σ((( つ◕ヮ◕)つ @ScanBcn]

@Teknium

[Nix0n @UnTalNixon_exe]

¿Sigues haciendo auditorías web a mano o perdiendo horas configurando herramientas genéricas para bug bounty? El juego en el hacking de aplicaciones web acaba de evolucionar.

La entrega número 26 de la serie sobre IA en ciberseguridad de Hackers-Arise desglosa el uso táctico de `Cybermes`, una herramienta impulsada por inteligencia artificial diseñada específicamente para romper aplicaciones web:

• Automatización letal: optimizada tanto para pentesting ofensivo como para bug bounty, agilizando el descubrimiento de vectores de ataque.
• Habilidades nativas: viene cargada con un set masivo de *skills* preconfiguradas para no perder tiempo montando flujos desde cero.
• Reportes listos para entrega: genera documentación técnica clara y estructurada de los hallazgos sin fricción.

Mientras la mitad de la industria sigue dependiendo de escáneres lentos y ruidosos, integrar agentes especializados en web hacking marca la diferencia entre encontrar el bug o quedarse mirando la pantalla.

Te dejo el artículo técnico completo por acá abajo para que lo analices y lo metas a tu stack de herramientas 🛡️👇

[Scan ─=≡Σ((( つ◕ヮ◕)つ @ScanBcn]

@cvander Prácticamente es mi barrio

[Scan ─=≡Σ((( つ◕ヮ◕)つ @ScanBcn]

@Arezno @JavierLobo Trump dijo "Aguántame el cubata" https://t.co/jdzmvaAPdu

[Scan ─=≡Σ((( つ◕ヮ◕)つ @ScanBcn]

@mild_sunrise https://t.co/ihjKPqbCd4

[Scan ─=≡Σ((( つ◕ヮ◕)つ @ScanBcn]

@mild_sunrise In Ukraine, they know a thing or two about that.

[GitHub @github]

Today is for the programmers 🍻

[Scan ─=≡Σ((( つ◕ヮ◕)つ @ScanBcn]

@enrub_en

[harley turan @hturan]

turns out 60hz hi-dpi eink screens feel pretty magical

[harley turan @hturan]

turns out 60hz hi-dpi eink screens feel pretty magical

[Luisma López @luisma_lopez]

Ante la duda, sigue el ejemplo de Indi.

[Techjunkie Aman @Techjunkie_Aman]

WhatsApp quietly rewrote its privacy story.

Not by removing encryption.

By changing everything around it.

12 updates that affect millions of users:

[International Cyber Digest @IntCyberDigest]

❗️Thunderbolt is finally coming to Linux on Apple Silicon thanks to volunteers who reverse-engineered it, without help from Apple.

Asahi's Sven Peter posted the first USB4/Thunderbolt driver for M1, M2 and M3 to the kernel list yesterday, after years of reverse-engineering.

[Bender El Que Ofende @BenderOfuscado]

Lo de Cantoná pateando a un fascista sí que era farmear aura. El resto, aficionados.