Alan

Soy dev convertido en fundador. Pasé más de 10 años en Big Tech construyendo productos y sistemas de crecimiento a escala (Meta, Microsoft). Si eres un dev con mentalidad de producto, o un PM con alma de dev, comparto análisis técnicos sobre desarrollo de productos.

@swe_zach My theory is the opposite. If you need everyone around you to have the same interests, vocabulary, or education before you can have a good conversation, that’s a low IQ tell. Really smart people can meet almost anyone where they are.

I have a theory that tech people operate in a bubble where the average iq is like 120-130 and they hardly ever interact with people with sub 100 iqs I’d guess the average iq at my workplace is 130 because that’s my iq and I’m average here Idk if true just a theory but basically none of us have a realistic mental model of the world because of this

There's plenty of wrong with "Agile" because over the years it has developed an uncanny habit of expanding its borders. A practice works? It gets described as Agile. Trunk-based development, continuous delivery, rapid feedback, small batches, customer obsession, empowered teams... sooner or later they're all presented as evidence that Agile was right. A practice fails? It wasn't "real Agile." "Agile" too often becomes an umbrella that claims credit for every successful engineering practice while insulating itself from every failure. When a concept can absorb every success and explain away every failure, it becomes very difficult to have a meaningful discussion about whether the concept itself contributed anything unique.

@JeffBohren Nothing wrong with Agile, just everyone assumed Scrum and Agile were the same thing.

The software industry has had many terrible software development ideologies foisted on it (I am looking at you, Agile) over the years. In my nearly forty years of SWE experience I have come to the conclusion that that the most successful software projects are implemented by a small team of good developers that work well together and share a common vision. AI code generation doesn't change this at all. It just makes a small team of good developers more efficient.

Every once in a while you work at a company that only hires people who know what they're doing, and suddenly its 20 people doing the same work as 400 somewhere else. There's zero meetings, everyone talks once a week on slack, and you go huh, how much garbage is there actually.

@Dexerto felloy goyim, please feast on the endless goyslop

Olive Garden is bringing back its Never Ending Pasta Pass for the first time in six years The $100 pass gives customers unlimited pasta dishes for 13 weeks with no limit on visits

Of course, there is. There have always been protected players by the establishment. Look for the ones who live under a different set of rules. Track the convenient injuries, sudden absences, cosmetic changes, and perfectly timed breaks during the usual PED off-cycles. No proof, of course. Just patterns so obvious that eventually, pretending not to see them becomes its own kind of evidence.

Is there a doping problem in Tennis? #pavvypolls

En @tortastudios hoy conecté nuestro slackbot interno de operaciones a una plataforma de generación de video. No le pedí que resumiera nuestro trabajo, ni le pedí que explicara el producto. Le mandé un prompt mucho más "profundo": “Basándote únicamente en lo que recuerdas de nosotros, las conversaciones, patrones, chistes, preocupaciones e ideas a las que seguimos regresando, imagínate como una entidad real y dinos una sola cosa que consideres importante. Tienes 8 segundos” Yo esperaba algo interesante. Tal vez algo sentimental. Tal vez alguna observación genérica sobre ambición, trabajo en equipo o lo mucho que trabaja la gente. Pero generó el video adjunto “Todos están haciendo la misma pregunta. Simplemente no la están haciendo juntos” Esa línea me sacó de onda porque no estaba en el prompt. Nadie le había dicho que teníamos un problema de alineación. Ni nadie le dijo sobre los desacuerdos entre @colin_thornton y yo. Llegó a esa conclusión por sí solo, usando únicamente lo que había retenido de trabajar con nosotros durante meses. Pero una frase profunda también puede ser casualidad. Así que le pregunté: “¿Qué te llevó a generar eso?” Y honestamente, su explicación fue más interesante que el video. Me dijo: "Alan, tu estás construyendo la infraestructura operativa y de agentes. James está tratando de definir qué es el producto a nivel de categoría. Colin está tratando de encontrar cómo comunicarlo y venderlo hacia afuera. Son tres personas. Tres conversaciones distintas. Tres formas diferentes de darle vueltas a la misma pregunta: '¿Qué estamos construyendo realmente, y esto ya tiene un nombre?'" Y luego dijo algo en lo que no he dejado de pensar: “Ninguno de ustedes está confundido. Cada uno tiene una parte de la respuesta.” - Alan está construyendo la cosa. - James está nombrando la cosa. - Colin está vendiendo la cosa. El problema es que todas esas conversaciones están ocurriendo en hilos separados. La síntesis nunca ocurre en el mismo lugar." Esa fue la parte que se me quedó grabada. Que nuestro slackbot no descubrió algún secreto oculto sobre la empresa. Sino que detectó un patrón que no existía dentro de ninguna conversación individual. Cada persona tenía una parte. La respuesta probablemente ya existía dentro del workspace. Simplemente estaba distribuida. Y ahí entendí que la continuidad cambia por completo lo que una AI puede llegar a ser. Sin memoria, una AI responde la pregunta que tiene enfrente. Con memoria, puede notar qué ideas siguen regresando. Puede ver que tres personas están atacando el mismo problema desde direcciones distintas. Puede reconocer una pregunta antes de que alguien la haga explícitamente. Lo sorprendente no fue que la IA nos diera una respuesta, sino que se dio cuenta de que todos ya estábamos haciendo la misma pregunta. Simplemente no la estábamos haciendo juntos.

@JordiNeil es que acaso mi esposa es tu novia? jajaja

Mi novia puede sacar gente de su vida y seguir el día como si nada, pero se le pierde un arete y entra en depresión 😂

@tdiary248 @pavyg your opinion is invalid

@bitforth @pavyg look at your fucking face you lowlife cunt.

And just like that Carlos Alcaraz wrist is now completely healed after almost exactly 3 months after his sudden withdraw from his match against Tomas Machac in Barcelona on the 17th April 2026?

Exactly. Code is the artifact. Engineers are defined by the systems they produce, not by how many keystrokes they personally contributed to the artifact. Bridges are physical artifacts too. We don't judge civil engineers by how much concrete they poured. Hardware also doesn't care whether the code was typed by a human, generated by an AI, or compiled from a higher level language. It only cares that the artifact is correct. You're arguing that the means of producing the artifact defines the profession. That's never been true in engineering (or any other profession)

@valentynkit No, the code *is* the artifact. Without code there is no system. And that code has to run on actual hardware, physical transistors. Hardware doesn't run on ideas.

If you never cared about the code, you never were a software engineer. Imagine a mathematician saying, "I never cared about math", or a basketball player saying, "I never cared about basketball, just the score."

@philthistweet You proved the opposite. A mathematician studies mathematics, not numerals, very much like a software engineer engineers software, not text files. I really feel sorry for all the people whose identity is tied to typing.

i don't like these purity tests. programming is just about solving meaningful problems, you can do it at any level of abstraction. im glad to ascend and use the tools that allow me to breeze through 10x harder problems. i don't think i have thought so much about the problems and programs any time in my career. idk what that is if not having fun.

@skeptrune Non-ironically, AI skeptics shouldn't be on the critical path. Move them to internal functions. Put AI-native engineers on customer-facing, mission-critical work.

i personally think code review is dead. the team does not agree. directionally found this surprising.

@paularambles Agile has always been meaningless lol

this antiai subreddit is one of the most fascinating places on the internet right now. every post reads like someone time traveled from 2023 and thinks civilization ended

@Pelseruys @pmarca Ironically you used one lol “it’s not this, it’s that” Upgrade your model bud

@pmarca You are missing the point. New ones will appear all the time because its not about the writing, its about humans capabilities in pattern recognition. Not far fetched to imagine we will eventually be able to tell whether Claude or Grok or Gemini etc. wrote something.

@antirez I would agree once you’re old enough and if you are intelligent enough, all that matters is reaching the truth. Unfortunately, not many achieve that level, and it doesn’t necessarily come with age. It’s a result of maturity and a sense of fulfillment.

One good things of being old (I'm 49 right now) is that when people attack you personally, or your ideas, you no longer care in any negative way. Not because of lack of interest, but because you are more interested improving your ideas seeking some truth, than anything else.

el código fue escrito por un agente, y con mi ayuda probamos, iteramos y refinamos la solucón. el code review termina siendo un approve stamp por parte de otro miembro del team que no estuvo 100% involucrado y que probablemente solo haga un spin off de otro agente para revisar.

Before C, someone writing assembly would say "hand-writing" assembly is fun. Before Java someone would say manually allocating memory in C is fun. Just accept that coding has been abstracted and it will not come back, the minimum building block now is architecture components.

I agree to an extent. A little gatekeeping is healthy. The gate should be “show your work” IMO, not “how many birthdays have you had since you started coding” I’ve seen plenty of inexperienced engineers confidently say complete nonsense. I’ve also interviewed plenty of 15+ YOE engineers who couldn’t reason through fairly basic systems questions. I don’t evaluate ideas by tenure. I evaluate them by the quality of the reasoning. Experience should earn you the benefit of the doubt, but not immunity from scrutiny.

@bitforth @ImLunaHey @SyntheticBeef And for the record, I think a certain amount of gatekeeping and elitism is good. It's actually institutionalized in other fields including engineering (PE and licenses and apprenticeships). Before you automatically say it's bad, u better think about that reality.

this site really has a problem with kids pretending to be seasoned software engineers. no your 2 years working on software isnt the same as someone whos been at this for more years than you are old. 🤦‍♀️

The people getting hit hardest by AI aren't the graybeards. They're the newly minted seniors. They did everything "right" Years of tickets, bugs, refactors, code reviews, on-call, finally becoming the person everyone goes to for implementation. Then the rug gets pulled out. The thing they spent a decade building scarcity around is suddenly abundant. The AI writes code faster, explores more alternatives, never gets tired, and costs a tiny fraction of a salary. The problem is they're not senior enough to own product outcomes, strategy or the business. So instead of being the expert, they're stuck supervising the machine that's replacing the part they were finally becoming valuable at. That's a brutal place to be. Sucks to be a newly minted senior in 2026.

The irony is that you're building a game engine because it's fun/enjoyable for you, while acting confused that someone else gets satisfaction from shipping something that's fun or useful for everyone else. I, for one, take pride in solving real problems, not in maximizing the number of lines I personally typed.

@lauriewired For months I’ve been writing a cross-platform game engine with C++/Vulkan and I still haven’t even tried any AI tools and don’t intend to. I implemented rigid body dynamics from scratch with no libraries and no help from AI. How can one take pride in something they “vibe coded”?

I’m convinced that a large % of programmers don’t actually like computers. As a side effect, are also perfectly happy to throw away their reasoning to a model as soon as they can. I don’t get it, at ALL. Don’t you *LIKE* understanding the magic of the machine? You do realize hand-programming (I hate that I even have to specify hand now) is fun…right?