Nicolas Ramirez

First trailer for the ‘HARRY POTTER’ series. Releasing this Christmas on HBO.

@koylanai @verge First time hearing Context Engineer. Is it software related?

Calling today's LLMs AGI, while the entire field is still built on careful context assembly and repeated behavioural correction, is just a drastic lowering of the bar. Current models are very good at interpolating across seen structures. But I still have not seen strong evidence of sustained, cross-domain, top-tier originality without substantial human framing, constraint, and correction. I'm a Context Engineer. My entire job exists because LLMs can't figure out what information they need. I spend 16 hours a day managing token budgets, debugging context degradation, and adjusting prompts word by word so models behave as expected. If AGI were here, my role wouldn't exist.

Nvidia CEO Jensen Huang says ‘I think we’ve achieved AGI’ theverge.com/ai-artificial-…

@Adidotdev In order to understand systems you need to at least understand how code works. Is not one or the other, you need both, and code is fondational

Hot take: The most valuable dev in 5 years won’t be the best coder… it’ll be the one who understands systems + users. Agree or disagree?

@AmanVirk1 I think you can add your own tables and schema to payloadcms V3, although I haven't tried

Are we (JS devs) using Admin panels? Laravel has Nova. Rails has Active Admin. Django ships one out of the box. What do we have 👀 Genuinely curious what everyone is using, or if this just isn't a thing in the JS world.

@mucotreso @AmanVirk1 That's an ORM, not an admin panel

@AmanVirk1 Maybe the equivalent in the JS world is Drizzle Studio or Prisma Studio. I know they're different, but that's what I see a lot of people using. Not sure if they're being used in production though.

@0xSero Bait

I had to open this abomination today, holy fuck. How can anyone use this?????

@5eniorDeveloper Pobres diablos jajaja

hahahahahaha

@sysxplore MacOS for work, Windows for gaming. Since I am using WSL Ubuntu on Windows I can get by without touching my Linux machine for development

Does anyone uses linux?

@HackingBaseball @johncrickett Loved the Quit your job bit 🤣

@johncrickett I did this as a goof: github.com/ImZackAdams/AI…

Everyone talks about how good AI agents are at writing code. But where's the actual software? Share your best example below.

Yes, and that can become expensive quickly. Are you doing only e2e tests? They take time to setup and run, environments isolated from production cost money. I have seen test suites take one hour+ to run. Performance testing can be even more troublesome. Are you going to invest in every feature the same amount of tests/care? Not every feature deserves that level of investing. Can you expand on what your approach would be?

@nicoramirezdev @edandersen It's such a common pitfall to talk about testing as if it's a 100% or 0% binary. No, testing is the only way you ensure a thing that worked today also works tomorrow. It's not a question of coverage %. Do you have a feature your customers expect to work? Test it! You have to!

Reading code, especially code you didn’t write, is 10x harder than writing code These people AI generating 90%+ of their code *are* reading it all, right… or are they just dumping the difficult verification work on their colleagues in PRs?

@tzhechev @edandersen Testing is always about tradeoffs. If you have a large and expensive test suite it is going to impact your development flow. It is not reasonable to expect 100% of edge cases to be tested 100% of the time. The test would require 10x more care than the source code.

@edandersen Reviewing code is massively overrated. Code either does what it should - which is testable - or it doesn't. Reading it doesn't provide a "more real" view of what it "really" does. I still think reviewing code is good, but it's a very limited quality assurance tool.

@FPupusas @TheVolun @ThePrimeagen Nah, reliance on multiple agents turn you away from your codebase in a way other factors would not. If you are ok with it, that's fine too

@TheVolun @ThePrimeagen I agree, just like overreliance on anything. doesnt mean we have to become luddites on a 6 month cycle yearning for the tech we were bashing last round

i am using supermaven again and i have something to say about this whole AI thing. I think as a group (swe) we rushed so fast into Agents when inline autocomplete + actual skills is crazy. A good autocomplete that is fast like supermaven actually makes marked proficiency gains, while saving me from cognitive debt that comes from agents. With agents you reach a point where you must fully rely on their output and your grip on the codebase slips. Its insane how good cursor Tab is. Seriously, I think we had something that genuinely makes improvement to ones code ability (if you have it). Truly acts as a multiplier, and we left it in the dust because it is not sexy. hurts me on the inside.

@AndrewYang How does abundance look like as an end state? Hard to build something without defining it

AI could flip us from scarcity to abundance if done right.

@asmah2107 What are you building that requires orchestration between agents? Genuinely curious

The most underrated architecture decision in multi-agent systems: Orchestrated (one boss agent delegates) vs. Choreographed (agents negotiate peer-to-peer) Microservices solved this in 2015. Nobody ported the answer to agents yet. The trade-offs are identical. The stakes are higher.

@devruso Por eso mismo, saben cómo protegerse de scrappers, saben cómo funcionan. Seguro don equipos diferentes en competencia, pensándolo bien.

la misma empresa que ofrece un servicio de protección contra scrapping ofrece un servicio para hacer scrapping. no elaboro mas

@mitchellh @thdxr Unless you are in a big corp, does it make sense to use AWS directly? For my hobby projects I get by with a VPS on Digital Ocean or Linode.

@thdxr AWS is full enterprise enshittification. Completely unusable for normal people nowadays, they've dropped the ball.

it took us several days to get our new AWS accounts properly provisioned with the right capacity limits to deploy our stuff it's crazy how things have regressed here

@marcelpociot I prefer to work in the terminal, so agents running on a tmux session makes more sense to me

Nobody can convince me that the CLI experience for AI agents is better than using a proper GUI 😅 That's just impossible

@PovilasKorop You mean beginner devs? I used to watch tutorials when I started to code...now It's more useful to read the docs. I guess newcomers don't see the value of that kind of content anymore

With AI, devs get quickly addicted to ready-made things. It's affecting my video numbers. Longer videos explaining the PROCESS and the WHY are losing. Shorter videos with ready-made tools/skills/prompts get highest views. Gonna go for a walk to think what to do with it.

Al final la IA terminó remplazando a los IA bros.

@iaparatodos25 @MoureDev Tu post parece un poco IA, para ser honesto

@MoureDev Titulares alarmistas: mal. Realidad: Code Review de Anthropic es una bestia multi-agente que revisa código IA-generated mejor que muchos seniors en bugs lógicos. No nos quita el puesto, nos quita el aburrimiento. Gracias por el toque de realidad 🙌

Me encanta Xataka y os leo desde hace media vida, pero con todo el respeto creo que a veces toca bajar un poco la intensidad de titulares como este. Lo siento mucho, pero estos mensajes generan una bola de nieve que solo deriva en miedo en el sector, abandono entre los estudiantes y desinformación entre quienes no programan, haciéndoles creer que ya no hace falta aprender nada para desarrollar software profesional.