Tobias Brandy

No te estoy ignorando, estoy juntando valor para contestarte

[Attractive @dailyatractive]

This goes to show once again that men's happiness is just that simple. I’d wager that 80% of the people in this group are strangers to one another.

[Sanya Timar @sanyatimar]

incredible take

[Jules Jacobs @JulesJacobs5]

The most fun in programming was the autocomplete era. Much better than the current slot machine + code review. It's a pity they stopped making better and better autocomplete models.

[pili 🪴 @0xpili_]

stop what you are doing and check this

https://t.co/LRbJQNmBAi

[Ceci Denot @gordameir]

Un medio inglés publica un artículo denunciando el apoyo militar que Israel le brindó a la Argentina durante la guerra de Malvinas y los guiños diplomáticos que le está haciendo hoy. En las respuestas hay muchos chillando que esto es “mentira”. Es increíble el nivel de desconocimiento que existe sobre este tema. Les guste o no, Israel fue el principal proveedor de armas a la Argentina durante la guerra de Malvinas: suministró desde abrigos, equipos de radio y combustible hasta aviones. Ingenieros israelíes incluso viajaron a las islas para enseñarles a los argentinos a operar los equipos de radio. Pero en Argentina circula mucha propaganda berreta como que “Inglaterra creó a Israel” o que “Israel siempre vota contra el reclamo argentino en la ONU” (ambas estupideces aparecen en las respuestas), entonces cuando ven a los propios británicos quejarse del apoyo israelí a la Argentina, en lugar de revisar un poco sus prejuicios reaccionan emocionalmente porque en sus cabezas Israel no puede estar asociado con nada positivo nunca

[John Cotterell @cotterzz]

I wondered what the dirac belt trick would look like with domain repetition.

[Glauber Costa @glcst]

I am excited to announce that we are officially writing a new version of Postgres. In Rust - and creating the LLVM of databases in the process.

In the span of a year, we have rewritten SQLite. Keeping the compatibility, increasing its feature set. MVCC, Types, (Live) Materialized Views, among other things. In the process of doing that, we have realized: At the end of the day, what makes SQLite special is that it compiles SQL to a database-specific bytecode. So why can't we compile *Postgres* to the same bytecode?

Turns out we can. I ran an experiment called pgmicro as a way to prove this hypothesis, and it works very well. It is time to make this official, and put the weight of Turso behind it. We shall give the world a modern take on Postgres. Wire compatible, but built on a new architecture.

We have already heard of others wanting to extend this. MySQL? Redis? the sky is the limit. What can we do if we do for databases what LLVM did for compilers? To prove how powerful the SQLite bytecode is, we are actually running DOOM compiled to the unmodified SQLite instruction set. And because Turso runs natively in the browser, you can play the game in your browser. With the database executing it.

Read the full story below! 👇

[fedez- @____fedez____]

Este es un mundial raro porque hay una instancia más, así que no está de más repasar las reglas:
1. Hasta cuartos, se ve el partido donde toque. Se paran las actividades durante el mismo. No hay celebración, aún estamos cumpliendo con la obligación.
2. Cuartos, ya no se va a laburar pero se gana y se sigue. Si es un rival muy importante, capaz un rato de cánticos en la calle.
3. Semis, todo suspendido y no se vuelve a las actividades. Si se gana hay obelisco light para los que viven más o menos cerca.
4. Final, paramos todo. No se velan familiares ni se realizan operaciones programadas. Si se pierde, obelisco un rato, para cumplir y mostrarse entero en la derrota. Si se gana, obelisco masivo, podés mamarte y romper infraestructura pública. Todas las actividades suspendidas hasta próximo aviso.

[pablo @fernandezpablo]

Inentendible lo de Scaloni no leyendo a los gordos de twitter, que advertían que Cabo Verde iba a ser difícil y que había que hacer muchos goles. A corregir para octavos.

[Fede’s intern 🥊 @fede_intern]

european football has spent the past fifteen years solving futbol like chess.

a generation of coaches optimized for pass completion, pressing triggers, territorial control, rest defense, and positional occupation.

the problem of this is that they optimize for what is measurable. depth, the willingness to attack space early, attempt the difficult pass, dribble past a defender, or deliberately create chaos, is a high variance play. it fails more often than it succeeds. if you evaluate players by completion rate, ball retention, or positional discipline, those actions look like mistakes. so they get coached out. eventually, everyone converges toward the same local optimum.

the game becomes increasingly legible. every team occupies similar spaces, presses in similar ways, builds from the back with similar patterns, and minimizes the same risks. systems become better at defeating other systems, but worse at dealing with players who refuse to behave like systems.

south american football never fully abandoned the duel as the fundamental unit of the game. the 1v1 remained sacred. so did the tactical foul, the unpredictable dribble, and the player willing to lose possession five times if the sixth breaks the match open. the objective was never simply to preserve structure, it was to create someone capable of destroying the opponent’s structure.

football is not won by completing the most passes. it is won by scoring more goals than the other team. those are related, but they are not the same objective.

this is the danger of optimizing proxies. when everyone optimizes the same measurements, they stop optimizing for victory itself. they optimize for looking efficient.

italy may have been the first major european football culture to lose part of its identity this way. its historical advantage was never athletic superiority or perfect positional play. it was tactical asymmetry, unpredictability, and an instinct for making matches uncomfortable. as italian football converged toward the same coaching model as the rest of europe, it gradually surrendered the qualities that had made it different.

the broader lesson extends well beyond football. every optimization process eventually risks becoming self-defeating. metrics become targets. proxies replace objectives. variance is mistaken for error. the outliers capable of breaking the system disappear because the system itself learns to eliminate them.

[Hank Green @hankgreen]

Nothing makes sense until you accept that what we find easy to pay attention to has very little connection to what is actually important.

[Brian Armstrong @brian_armstrong]

How to keep AI spend flat while token usage grows exponentially: Not with friction and spend alerts. With better defaults, routing, and caching.

Better Defaults (not Usage Caps) – Engineers can choose any model they want, but defaults matter. We’re experimenting with defaulting to open weight models like GLM 5.2 and Kimi 2.7 through our LLM gateway, while still encouraging engineers to choose the right model for the task. 91% of our employees were never hitting their usage caps, so instead of lowering caps and driving up alerts, we're moving to cheaper defaults. Note that code reviews use a diversity of models, so they can check each other's work.

Better Routing – In our custom harnesses, we preprocess prompts and route to the best model for the job, considering cache hits and model pricing. For instance, you may want a frontier model for planning, but not for execution where they can be overkill. Ultimately, humans shouldn't be choosing models - AI can automate this task.

Better Caching – Cache misses are the easiest way to drive your cost up. All of our requests are cache aware, so we’re reusing a warm cache wherever possible. For example, our cache hit rate went from 5% → 60% in LibreChat once properly implemented.

Keep Context Lean – Start fresh sessions when switching tasks. Scope file context narrowly. Disconnect unused tools. Don't just compact. The goal isn't fewer tokens used, it's fewer tokens wasted.

Better Visibility – Our engineers can use as many tokens as they want, from whatever model they want, but we’ve made usage visible – and the more you spend on AI, the more impact we expect.

The goal isn't to suppress usage. It's to build the infrastructure that makes exponential growth sustainable.

Putting this into practice has cut our AI spend nearly in half, while our token usage continues to grow.

[soli @solisolsoli]

By Sarah Jacoby

[Tobias Brandy @TobiasBrandy]

https://t.co/SigD4nVWp4

[Tobias Brandy @TobiasBrandy]

https://t.co/NrEom87W24

[atlas @creatine_cycle]

the elites don't want you to know this but the biggest biohacking anti-aging hack is living in walking distance to your friends