I care what the code looks like because I care about the portfolio of options embedded in the code. If I assume that the first thing I ship will deliver all of the value, then sure nobody cares about the internals. If I'm going to learn & expand, though...

The problem with the "if it works who cares what the code looks like" mindset for agentic work is that it assumes the agent has a perfect understanding of "works." Realistically, things are underspecified, agents make bad assumptions, etc. To be fair, agents are pretty good at unit test coverage. They're pretty bad at designing human experiences (API, CLI flags, etc.), especially cohesive ones for future roadmap plans they may not have visibility into (unless your backlog is perfect and vision fully laid out, which I doubt). They're bad at knowing where performance matters and what type (CPU vs memory tradeoffs). They're bad at where compatibility matters and where it doesn't (and tend to err on the side of preserving it without further guidance). Etc. Unless you have this ALL specified, you can't possibly claim "it works" without taking a look and thinking about it.

More AI-generated code doesn't make your team faster. It might actually slow you down.

I was recently diagnosed with a rare autoimmune disorder called anti-NMDA receptor encephalitis. It's a "disease of chaos" that completely upturned my life for a couple months. I wrote a blog about it that goes into more detail and discusses prognosis. burntsushi.net/encephalitis/

Nowadays everyone is saying "The role of developer is changed. The focus is no more on programming language, syntax, libraries, etc. Developers should focus on architecture and how system works". It's amusing. Now, take a junior developer who never done much of hands-on coding. Give them access to Claude/Codex or whatever and tell them to focus on "Architecture". What could possibly go wrong!!! I think "we" will never learn NOT to go to extremes with every new idea.

Deep inner suffering inevitably arises when the human person is reduced to performance, consumption, or a statistical datum. Many young people today live under the yoke of expectations to perform, immersed in an exasperated competitiveness that generates anxiety, fear of not measuring up, and disorientation.

Artificial intelligences do not undergo experiences, do not possess a body, do not feel joy or pain, do not mature through relationships, and do not know from within what love, work, friendship or responsibility mean. Nor do they have a moral conscience, since they do not judge good and evil, grasp the ultimate meaning of situations, or bear responsibility for consequences. They may imitate or even simulate, but they do not understand what they produce, for they lack the affective, relational, and spiritual perspective through which human beings grow in wisdom. #MagnificaHumanitas

The industry has gone completely nuts. Use tokens to generate AI code and documentation slop. Then use even more tokens to understand and review that slop. Then judge engineers by token usage instead of how empathetic and clear their docs and code actually are, and completely neglect human comprehension. Utter nonsense.

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem. As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)! I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work. It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results? 88ms => 1.5ms 150K allocs => ~500 allocs Incredible right? Nope. My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path. This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput. The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity. Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

Murió Carmen Teresa Navas, madre de Victor Hugo Quero Navas, preso político desde enero 2025, que fue detenido arbitrariamente, desaparecido, torturado y muerto bajo custodia del Estado. Por meses las autoridades de distintas instituciones, de forma coordinada, se burlaron de su madre y la hicieron peregrinar buscando a su hijo. Hace 10 días lo encontró y estaba en una fosa. Hoy ella ha muerto. Tener un familiar preso es también descapitalizarse, privarse de atención médica, medicinas, comidas y vivir en ansiedad, con los sentidos alterados, siendo revictimizados con cada visita a instituciones, con cada petición de justicia y con cada segundo de esperar una llamada telefónica para saber algo nuevo. En enero de este año también hubo varias muertes de madres esperando liberaciones en esa semana en la que Jorge Rodríguez decidió mentir otra vez y decir que todos serían liberados, cuando ahora, cinco meses después, van apenas por la mitad y son decenas más los que están desaparecidos, como desaparecido tuvieron a Víctor Hugo Quero. El Estado es responsable de todo esto. Son crímenes de lesa humanidad que generan un daño atroz más allá de las víctimas directas y sus familiares. Esto nos rompe a todos y por eso también es trabajo de todos exigir justicia y no olvidar nada de lo que ocurre, porque los pactos de silencio y los reacomodos con la dictadura solo benefician a la impunidad y multiplican los crímenes. El mensaje más importante que ha enviado el chavismo con la muerte de Victor Hugo Quero, como con todas las muertes bajo su custodia, es que siempre pueden hacerlo contra cualquiera. Se lo hicieron al compadre de Hugo Chávez, el general Baduel, se lo hicieron a Fernando Albán, y se lo pueden hacer a quien necesiten para imponer el miedo y la paralización. Por eso es impresionante que los venezolanos en general no hayamos tirado la toalla y mantengamos las exigencias de justicia, democracia y libertad. Para trascender el dolor, para ser justos y para reclamar lo que nos pertenece. El secuestro institucional que permitió la muerte de Victor Hugo Quero, que se burló de su madre y que sigue negando el paradero de decenas de presos, tiene que acabarse por completo.

La declaración del ministerio sobre la muerte de Victor Hugo Quero Navas no es consistente. La señora Carmen Teresa Navas tiene mucho tiempo buscando a su hijo y por meses la ignoraron, la presionaron y incluso fue sometida a interrogatorios revictimizantes. Los defensores de derechos humanos y los medios han apoyado este caso y le han dado mucha visibilidad. El Estado sabía que estaba muerto desde el principio, ¿Entonces por qué no lo informaron antes? ¿Cuáles son los funcionarios y las cadenas de mando responsables tanto del ocultamiento de la información como las negativas en cada centro de detención y oficinas públicas a las que acudieron a preguntar? ¿Quiénes eran los funcionarios responsables de la custodia, la salud y la integridad del centro donde estaba preso? ¿Quiénes eran sus jefes? ¿Quiénes lo detuvieron en enero de 2025 y por qué? ¿Por qué le impidieron nombrar a su abogado de confianza? ¿Por qué le negaron comunicarse oportunamente con su madre en todos sus meses de presidio? ¿Por qué no se han investigado aún las denuncias de torturas y malos tratos en los centros de reclusión? Los presos no tienen hemorragias internas de forma espontánea. ¿A qué organismos internacionales, independientes y confiables pueden llamar para que establezca la verdad de lo ocurrido? Es absurdo que se le deje esa responsabilidad a la misma gente que ocultó por tanto tiempo este y otros casos, como el de Hugo Marino y tantos otros. ¿Se dan cuenta de lo grave que es la desaparición forzada y todo lo que trae como consecuencia? ¿Quiénes van a proteger a quiénes para garantizarles impunidad?

Normalize disappearing for an extended period of time to do a factory reset on your focus and creativity.

I’ve been harping on the disciplines and tools for using AI lately. I find them to be a very effective approach. But I don’t want to leave you with the impression that a few simple disciplines and tools is sufficient. As the AI’s build software, you — the software engineer — need to have a good mental model of what the AI is doing. You need to apply engineering insight to correct it when it takes a path you don’t like. You have to be an active manager in the design and architecture of the system. You have to be able to “see within“ without resorting to exhaustive code reviews. You have to form suspicions about what the AI is doing, and you have to probe and experiment to verify your suspicions.

The impact and ramifications of this tweet, will hopefully start conversations around being reliable more and more. I hope this is *the* wakeup call for the tech industry!

A lot of my bros have been running into Venezuela since it opened up. I myself have a visa pending to go visit. That said, things don’t change overnight. This week a friend of mine was visiting her hometown (she lives in USA now) and the local gangs found out she was in town. At 1 am, 16 men pulled up to her house, broke in, pulled her out of bed and took her to an undisclosed location trying to extort money out of her. Guns to her head, threatening to rape her. They held her though the night and released her the next morning unharmed when she finally convinced them she had no money. Fuck. That.

While AI agents make building software much faster (esp for experienced devs) - they seem to not make it any easier for early-stage startups to find PMF. Talked with 2x “AI-pilled” founders who are v productive devs, are building their startups. It remains damn hard, AI or not

AI use in offices is reportedly producing “workslop” that is taking more time to fix than to create from scratch, per BetterUp Labs

Dario demonstrating that he doesn't understand software engineering. The human side of what we do has always been the heart. I can empathize with *wanting* to never talk to an engineer ever again, but engineering becomes more important with better tools.

The only people who believe any of this are non-coders. I tried to build a game (an area I’m an n00b in.) The results are amusingly disastrous - I never before coded a decent game. But I’ll crack out backend services w AI rapidly - because I coded dozens of them before…

Unpopular opinion: "AI makes everyone a developer" is true the same way "cameras makes everyone a photographer"