Hoy publico mi primer libro: «Menos software, más impacto». Da un poco de vértigo.
La tesis: tu equipo no va lento por escribir mal código. Va lento porque escribe código de más. 🧵
Este hilo es para parar un momento, respirar y dar las gracias. En 72h todo esto ha escalado a un nivel que nunca imaginé: prensa, más de 10k recaudados, mensajes de oncólogos de todo el mundo y la comunidad tech volcada con ideas para usar IA en cáncer.
Empecé pidiendo ayuda para un caso ultra raro de cáncer de mama metastásico, pensando que igual alguien, en algún sitio, podría interesarse por investigarlo. Lo que ha pasado después demuestra que las redes, usadas con responsabilidad, pueden abrir puertas reales entre pacientes, oncólogos e investigadores.
La comunidad tech y de IA se ha tomado esto muy en serio: gente proponiendo modelos, creando repos de apps, para hacer accesible la información compleja. De aquí van a salir ideas muy interesantes para quienes vengan detrás de mí.
Quiero que quede claro: todo esto no va solo de mí. Va de que los casos raros dejen de ser notas a pie de página y pasen a ser datos, estudios, opciones reales. Cada euro donado y cada retuit está empujando esa idea. Gracias, de verdad.
En la bio tenéis el enlace para quien quiera seguir apoyando la parte de investigación y pruebas avanzadas de mi caso. Y, sobre todo, seguid haciendo ruido inteligente: preguntando, compartiendo información buena, conectando ciencia y personas. Eso también salva vidas.
Ahora viajo a Madrid a seguir avanzando con todo esto, se vienen cosas increíbles.
GRACIAS 💜
Tengo 35 años y cancer de mama metastásico, un caso raro, menos del 1% de tumores de mama son como el mío y hay poca documentación sobre ello.
Por eso me gustaría encontrar personas que se dediquen a esto y que quieran investigar con mi caso. Twitter haz tu magia
the answer is rather simple.
"hey this tech is super cool, look at what i can do!" - agreed. it is pure magic that as i write this twitter post, my 100% vibe coded app continues to get features. Thats fun.
"hey, you better be producing thousands of lines of code a day and this is the only normal that will be accepted in the future" - this is a completely different message.
i think most peoples problem with gary is that he frames things in such crappy terms. I would be completely down to celebrate the Ws if it was the former, but it is instead the latter. And i am tired boss. I am tired of getting hacked on every website and having every website run like shit.
right now everything in the world is telling you to go faster, ship more, add that feature, start another project
so i'm actively working on feeling ok not doing any of that
Dirty industry secret: nobody really knows the best way to use AI for software development 📢
- 24 months ago, we copy-pasted from ChatGPT
- 18 months ago, we jumped between ask mode and agent mode
- 12 months ago, we told AI "you are a senior developer"
- 9 months ago, we built MCPs
- 6 months ago, we switched to plan mode
- today, we're obsessed with skills
All of this (and much more) are just early experiments and temporary hacks in a very young and quickly evolving field.
So when someone says their workflow is the optimal one, they're confused at best.
Stay curious, stay open, stay in control and stick to the fundamentals, and you'll come on top in this amazing tech revolution. Enjoy the ride!
I give it less than 6 months before Garry stops preaching LOC and starts preaching maintainable code bases.
And with that one move he will go from junior engineer to a bit more senior.
We watching his Eng journey live 🍿
Salario Medio Vs Alquiler Medio 📊
🇨🇭 Suiza
6.500€ / 2.100€ → 32%
🇩🇰 Dinamarca
4.200€ / 1.400€ → 33%
🇳🇴 Noruega
4.400€ / 1.500€ → 34%
🇳🇱 Países Bajos
4.000€ / 1.500€ → 37%
🇸🇪 Suecia
3.200€ / 1.300€ → 41%
🇫🇮 Finlandia
3.300€ / 1.400€ → 42%
🇮🇪 Irlanda
3.800€ / 1.600€ → 42%
🇩🇪 Alemania
3.000€ / 1.300€ → 43%
🇫🇷 Francia
2.800€ / 1.300€ → 46%
🇪🇸 España
1.600€ / 1.100€ → 69%
En España casi el 70% del sueldo medio se iría solo en alquilar. Luego nos preguntamos por qué cuesta tanto ahorrar.
¿Que opináis? Os leo 👇
Lying is always a bad way to start a relationship: a job, a customer, a spouse.
Even if it "works,” the other person fell in love with someone else, something else. A constant burden on you, and will be pain until it ends.
Not the way to live.
siento que con todo el tema de la IA se ha creado una expectativa brutal de tener que producir más. si no tienes 20 agentes cerrándote tickets, te vas a quedar atrás. si no estás vibecodeando 7 proyectos a la vez, te vas a quedar atrás. si revisas el código estás perdiendo tiempo y, por lo tanto, te vas a quedar atrás. si tu empresa no tiene capex x20 dedicado a IA, se va a quedar atrás. todo debe ser creado a un ritmo cuestionable que hace que el propio equipo dentro de la empresa no sea capaz de asimilar lo que está ocurriendo en la empresa; el propio consumidor o cliente final no sea capaz de asimilar las opciones que se le están dando, no hablemos de dar algún tipo de feedback para que el ciclo natural de mejora de producto pueda tener su lugar.
hemos pasado de iterar las cosas a actuar como auténticos babuinos en celo, lanzando mierda al input de un pipeline robotizado que a su vez escupe por su output la misma mierda en distinta forma.
“he vibecodeado un compilador de c!!!”
pero espera, ese compilador falla al intentar compilar un simple “hola mundo” y, cuando no falla, produce unos binarios extremadamente lentos. esto es inservible en la práctica.
“bruh, no entiendes, esto es el progreso!!! y además esto ya es old news. ahora he vibecodeado un clon de next.js!!!”
pero espera, ese clon está lleno de errores críticos de seguridad. esto no se debería usar en absoluto.
“bruh, es que no importa que sea malo ahora!!! será bueno en el futuro!!!!!!!!”
o no, pero aunque lo fuera, estas usando lo actual (que tiene problemas) en producción. estás enshitificando todo.
la calidad, en general, de todo, está bajando a un ritmo que da miedo.
hemos pasado de exigir que las cosas funcionen bien, rápido y de manera eficiente a conformarnos con half-baked scripts que se unen con superglue con otros scripts, para poder ir más rápido.
¿pero ir más rápido a dónde? ¿a donde estamos yendo? ¿que cojones estamos haciendo?
Cons of being a software engineer no one really talks about 👇
Everyone sees the salaries, WFH perks, and fancy titles.
Here’s the other side:
• Your learning never ends
If you stop learning for even 6–12 months, you start falling behind. Tech doesn’t wait.
• Mental fatigue is real
Staring at screens, debugging for hours, context switching—your brain gets tired before your body.
• You’re judged by output, not effort
10 hours of hard thinking can look like “nothing” if the feature didn’t ship.
• Imposter syndrome never fully goes away
New stack, new company, smarter peers → constant self-doubt.
• Deadlines don’t care about bugs
Management promises dates. Engineers deal with reality.
• Work-life balance depends on the team, not the role
One bad manager can ruin a “dream job”.
• Rejections hit differently
You can be good and still fail interviews because of luck, timing, or niche questions.
• Side projects feel mandatory
To grow or switch jobs, your free time often becomes “resume time”.
• AI pressure is increasing
You’re expected to be faster, smarter, and more productive—constantly.
• Your worth can feel tied to your skills
When code breaks, confidence breaks with it.
Still a great career.
Just not the “easy money” path social media sells.
If you’re choosing this field—choose it with eyes open.
Bit by bit, we are starting to see what the new AI assisted software development world is going to look like for the next several years.
My current (still evolving) take:
- Massive unleashing of experimental work, proofs of concept, rough drafts
This should lead to a huge boost in the amount and creativity of software products that come to market, at the cost of a sudden increase in noise, a veritable din
- Significant decline in average code quality
Some code gets better, a lot gets worse, and the limitations of the current technology and unlocking less experienced developers to create software will lead to a near crisis in poorly built products in the near term
- Large proliferation of tools
As we scramble to adapt, experimentation and perspectives will lead to a vast array of possible solutions, each with their own sets of tradeoffs. Reminds me of the early days of Web 2.0, where there was a new framework every week and they all sucked
- Some emerging best practices
Over the past 6 weeks I have talked to many very experienced developers (often 1 on 1 video calls), and we are now starting to circle some common threads for AI-assisted software dev best practices (these are off the cuff, so don’t expect perfection):
1. Slow down, learn the tools, figure out the tradeoffs
2. Quality still matters when it matters; often, the existing models and tools fall short of maintaining that quality on larger code bases
3. The developer is responsible for the code they ship
4. Documentation (via skills, tasks, or just markdown docs) is tremendously helpful, but you should also understand it, not just rely on the AI to
5. High level architecture is still an area where a human with a lot of experience can add a ton of value
6. AI is not a substitute for good taste (and caring about things)
7. Some techniques are locally productive and globally harmful (more on this below)
8. The bottleneck is in review, understanding, and higher level systems architecture more so than coding speed.
9. Some developers are more adept than others at various parts of this new value chain. Current teams are full of developers vetted for and hired to do one job who are facing a significantly different way of doing it.
10. Coding itself might get done a different way, but the fundamental engineering patterns are often still extremely important. Not in cases where it’s just about satisfying some developer love of symmetry, but definitely in domains like data modeling and the like.
More on local optimization vs global concerns: agent chains like Ralph and aggressive code gen can feel incredibly fast, but tend to accumulate inconsistencies and tech debt over time. Speed at a file/feature level does not guarantee speed at the overall system level, and I have felt this personally when I’ve leaned too hard on such tools.
- We will learn more as time moves on. Be kind
We are adapting, evolving, playing with the tools, sharing, feeling the bruises when we get it wrong. There are educators trying to stay ahead of it and provide value. There are normal devs just trying to make a living, and stay relevant. None of this is as unique to you as you might think — I hear from others, and they’re feeling the same pressures. Be kind to each other
A few random notes from claude coding quite a bit last few weeks.
Coding workflow. Given the latest lift in LLM coding capability, like many others I rapidly went from about 80% manual+autocomplete coding and 20% agents in November to 80% agent coding and 20% edits+touchups in December. i.e. I really am mostly programming in English now, a bit sheepishly telling the LLM what code to write... in words. It hurts the ego a bit but the power to operate over software in large "code actions" is just too net useful, especially once you adapt to it, configure it, learn to use it, and wrap your head around what it can and cannot do. This is easily the biggest change to my basic coding workflow in ~2 decades of programming and it happened over the course of a few weeks. I'd expect something similar to be happening to well into double digit percent of engineers out there, while the awareness of it in the general population feels well into low single digit percent.
IDEs/agent swarms/fallability. Both the "no need for IDE anymore" hype and the "agent swarm" hype is imo too much for right now. The models definitely still make mistakes and if you have any code you actually care about I would watch them like a hawk, in a nice large IDE on the side. The mistakes have changed a lot - they are not simple syntax errors anymore, they are subtle conceptual errors that a slightly sloppy, hasty junior dev might do. The most common category is that the models make wrong assumptions on your behalf and just run along with them without checking. They also don't manage their confusion, they don't seek clarifications, they don't surface inconsistencies, they don't present tradeoffs, they don't push back when they should, and they are still a little too sycophantic. Things get better in plan mode, but there is some need for a lightweight inline plan mode. They also really like to overcomplicate code and APIs, they bloat abstractions, they don't clean up dead code after themselves, etc. They will implement an inefficient, bloated, brittle construction over 1000 lines of code and it's up to you to be like "umm couldn't you just do this instead?" and they will be like "of course!" and immediately cut it down to 100 lines. They still sometimes change/remove comments and code they don't like or don't sufficiently understand as side effects, even if it is orthogonal to the task at hand. All of this happens despite a few simple attempts to fix it via instructions in CLAUDE . md. Despite all these issues, it is still a net huge improvement and it's very difficult to imagine going back to manual coding. TLDR everyone has their developing flow, my current is a small few CC sessions on the left in ghostty windows/tabs and an IDE on the right for viewing the code + manual edits.
Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later. You realize that stamina is a core bottleneck to work and that with LLMs in hand it has been dramatically increased.
Speedups. It's not clear how to measure the "speedup" of LLM assistance. Certainly I feel net way faster at what I was going to do, but the main effect is that I do a lot more than I was going to do because 1) I can code up all kinds of things that just wouldn't have been worth coding before and 2) I can approach code that I couldn't work on before because of knowledge/skill issue. So certainly it's speedup, but it's possibly a lot more an expansion.
Leverage. LLMs are exceptionally good at looping until they meet specific goals and this is where most of the "feel the AGI" magic is to be found. Don't tell it what to do, give it success criteria and watch it go. Get it to write tests first and then pass them. Put it in the loop with a browser MCP. Write the naive algorithm that is very likely correct first, then ask it to optimize it while preserving correctness. Change your approach from imperative to declarative to get the agents looping longer and gain leverage.
Fun. I didn't anticipate that with agents programming feels *more* fun because a lot of the fill in the blanks drudgery is removed and what remains is the creative part. I also feel less blocked/stuck (which is not fun) and I experience a lot more courage because there's almost always a way to work hand in hand with it to make some positive progress. I have seen the opposite sentiment from other people too; LLM coding will split up engineers based on those who primarily liked coding and those who primarily liked building.
Atrophy. I've already noticed that I am slowly starting to atrophy my ability to write code manually. Generation (writing code) and discrimination (reading code) are different capabilities in the brain. Largely due to all the little mostly syntactic details involved in programming, you can review code just fine even if you struggle to write it.
Slopacolypse. I am bracing for 2026 as the year of the slopacolypse across all of github, substack, arxiv, X/instagram, and generally all digital media. We're also going to see a lot more AI hype productivity theater (is that even possible?), on the side of actual, real improvements.
Questions. A few of the questions on my mind:
- What happens to the "10X engineer" - the ratio of productivity between the mean and the max engineer? It's quite possible that this grows *a lot*.
- Armed with LLMs, do generalists increasingly outperform specialists? LLMs are a lot better at fill in the blanks (the micro) than grand strategy (the macro).
- What does LLM coding feel like in the future? Is it like playing StarCraft? Playing Factorio? Playing music?
- How much of society is bottlenecked by digital knowledge work?
TLDR Where does this leave us? LLM agent capabilities (Claude & Codex especially) have crossed some kind of threshold of coherence around December 2025 and caused a phase shift in software engineering and closely related. The intelligence part suddenly feels quite a bit ahead of all the rest of it - integrations (tools, knowledge), the necessity for new organizational workflows, processes, diffusion more generally. 2026 is going to be a high energy year as the industry metabolizes the new capability.
I have to be honest. It's really hard to not panic in this atmosphere. If you're scared where this is all going, yeah me too. I personally worry too much about being irrelevant. I know it will all work out. Keep calm and Vibe on.
28 months in to 6 months from AI taking your jobs
* 4 months into 24 months until cursor is obsolete
* 6 months into 6 months until ai writes 90% of your code (part 2, the codening)
Creo que los más nuevos no pueden entender lo que significaba la LVP en su día.. No había nada mejor, aún recuerdo la alegría al recibir el mail en el que nos decían que estábamos dentro en 2014. Es una pena que todo haya llevado a esto..