Paulo Chaves

Senior Software Engineer | TypeScript, Node.js, React.js

[Paulo Chaves @opchaves]

that's true. knowing when what you have is enough is precious.

[Maximilian @maxedapps]

Let me repeat: Agents will ALWAYS find more issues when you ask them for it.

You can have them write the code and let different agents with fresh context review that same code and the "improved code" dozens of times => they will not stop finding issues.

It's YOUR job to determine which issues really are issues. And which potential improvements you really want or need.

[Paulo Chaves @opchaves]

@thdxr I'm building a pretty good one. https://t.co/db733S28JN

[Paulo Chaves @opchaves]

Yeah, with coding agents you produce way more code in much less time but you still have to test, which is obvious, if you really care about the product. And testing still takes some time.

[Paulo Chaves @opchaves]

@Semper_Viventem I keep this setting on for transparency. I want to make it clear that Claude wrote the code. I used to turn it off. I’’m glad CC is helping me :)

[Paulo Chaves @opchaves]

All I need now is crafting a great prompt with the things I need and how I want them to work. No need to worry about tech stack. Just focus on the app itself. I shall not let AI guess what I want.

[Paulo Chaves @opchaves]

I can be build a really cool mobile app that has a real backend with just a single prompt. That app comes with great UI, database and RLS policies all set up.

[Paulo Chaves @opchaves]

Same words for us humans, but LLMs sees them differently depending on case and position. GPT-5.x

“ man” = 873

[Paulo Chaves @opchaves]

IA estudar, agora estudo IA

[Paulo Chaves @opchaves]

@eduantao ciclismo

[Paulo Chaves @opchaves]

@jarredsumner Great!
I wander how many agents running in parallel were needed to make this happen…

[Paulo Chaves @opchaves]

@RealGalego Está mais vivo do que nunca. Claudim está sempre disponível

[Paulo Chaves @opchaves]

Prompt: make it easy to reproduce this issue. do a temp code change if needed 🫡

[Thariq @trq212]

https://t.co/MXt5XS4xBX

[Chris Wood @CWood_sdf]

i love how people are saying "if we write a sufficiently detailed specification, the agent can write all our code"

do you know what writing a sufficiently detailed specification that deterministically maps to what a computer's actions is? it's coding

[Anthropic @AnthropicAI]

New Anthropic research: Natural Language Autoencoders.

Models like Claude talk in words but think in numbers. The numbers—called activations—encode Claude’s thoughts, but not in a language we can read.

Here, we train Claude to translate its activations into human-readable text.

[AshutoshShrivastava @ai_for_success]

SpaceXAI + Anthropic deal in a nutshell.

[liamtran @liamtrn]

@ClaudeDevs thank team, great work!!

[Paulo Chaves @opchaves]

Opencode Go is really interesting. I will definitely check it out.

[Drizzle ORM @DrizzleORM]

Drizzle v1.0.0-rc.1 is out 🚀

▪︎ Effect v4 native support
▪︎ JIT row mappers to reduce ORM overhead to ~0
▪︎ Reworked casing API (breaking change)
▪︎ Drizzle for LLM agents (preview)

Drizzle is now as fast as using raw driver and mapping(or not mapping) results by hand 🙃

[Paulo Chaves @opchaves]

Good old days

[Dmitrii Kovanikov @ChShersh]

I miss the toxic tech culture of receiving 50+ code review comments.

Now everyone just pushes AI slop and doesn’t care as long as CI is green.