Software Engineer | Independent Researcher | Edge Computing | AI Agents

Madrid, Spain
Joined January 2021
Hey, jovem! Estou cada vez menos utilizando redes sociais para publicar coisas e ficando mais passivo no rolê. Porém eu compartilho os artigos que leio e outros links no meu LinkedIn (sem postagem de motivação tosca) Me segue lá linkedin.com/in/william-oliv…
1
8
1,655
William Oliveira (bsky.app/profile/1ilhas) retweeted
Why agentic swe is a failure. A tool that turns humans into a bottleneck is utterly ill designed from the very start. We need tools that make humans better at what they do best, not tools that turn us into slaves.
Anthropic’s report is basically saying humans are the bottleneck. “Amid this model-driven acceleration, human attention is a constant bottleneck. Claude sessions amount to around 80% idly waiting for human input. As one engineer put it, people are “bottlenecking the agents.”
16
15
1
174
3,542
William Oliveira (bsky.app/profile/1ilhas) retweeted
People after every new model…..
62
73
9
842
26,636
William Oliveira (bsky.app/profile/1ilhas) retweeted
safado, eu sei o que vc quer com isso
482
3,162
223
21,382
153,531
William Oliveira (bsky.app/profile/1ilhas) retweeted
I miss Comic Sans.
213
18,422
333
268,766
1,708,962
William Oliveira (bsky.app/profile/1ilhas) retweeted
Opus 5.5 made gummy bears. Gemini 4 Argon made gummy statues Same prompt. Opus on top, Gemini below I put both through the same moves - pull a paw, yank the head, squish it flat, throw it, then pour a whole bag on top Opus bears are actual jelly. Pull a paw and it stretches into a thin rubbery strand, then snaps back and wobbles. Squash one and it pancakes, then launches itself back up. Throw it and it bounces off the floor and shoves its friends out of the way Gemini bears are solid plastic. The arm turns into a stick, the bear flies straight through its neighbor, and the pile at the end just clips into itself And Opus nailed the look too - translucent, glossy, light passing right through them Opus: $7.2 Gemini: $22
Readers added context they thought people might want to know
Multiple users reproducing the prompt with Gemini 4 Argon (medium/high) obtained soft deformable jelly physics like the top panel, not rigid statues. x.com/caxvalencia/st… x.com/Alxander1024/s… x.com/Ufdcqg/status/…
212
234
146
3,984
626,552
William Oliveira (bsky.app/profile/1ilhas) retweeted
8
112
1
3,596
28,037
William Oliveira (bsky.app/profile/1ilhas) retweeted
Como disse o @NPTO, é uma gente rica, que quer ficar mais rica. É uma gente rica, que não quer mais ninguém rico. É uma gente rica, que não quer que seu filho estude e passe na frente do filho burro dela. É uma gente rica, que é mais rica do que gente.
2
1,401
61
6,056
73,806
William Oliveira (bsky.app/profile/1ilhas) retweeted
“A esquerda só sabe performar” Bolsonarista mais fraco:
49
1,634
31
11,867
71,260
William Oliveira (bsky.app/profile/1ilhas) retweeted
28
2,724
119
13,838
99,739
William Oliveira (bsky.app/profile/1ilhas) retweeted
começaram a decompilar software proprietário tipo o pacote office e suíte adobe e eu acho isso incrivel pq quando eram só pessoas comuns tendo seu trabalho roubado por IA ninguém dava a mínima mas agora q tão roubado trabalho de grandes empresas o negócio vai ficar bem engraçado
67
889
75
8,285
186,427
William Oliveira (bsky.app/profile/1ilhas) retweeted
😜
20
3,331
23
14,660
73,113
William Oliveira (bsky.app/profile/1ilhas) retweeted
Replying to @lucas_builds
I recommend CDD: Customer-Driven Development. Let your clients find the bugs. If they don’t, it’s not a bug 😶
55
69
16
3,735
100,948
William Oliveira (bsky.app/profile/1ilhas) retweeted
why do people want Adobe to lose to AI so badly? i mean
264
384
45
8,942
123,719
William Oliveira (bsky.app/profile/1ilhas) retweeted
Consciência de classe na sua mais completa tradução...
263
15,521
336
119,330
1,032,823
William Oliveira (bsky.app/profile/1ilhas) retweeted
Today we are announcing our biggest bet yet on our own silicon. Starting next calendar year, we will scale our Maia family to multi-gigawatt deployments, making it the primary inference system for our own product and services. And we’re introducing Maia 300, which delivers 40% more tokens per watt compared to leading GPU-based systems. We are already deploying Maia 200 across the US, Europe, and Australia, delivering up to 1.7x more tokens per watt across experiences in Teams and Copilot. Our goal is to build on the full-stack efficiency gains we’ve already seen, while continuing to bring the latest innovations from across the ecosystem into our fleet.
794
1,632
613
15,670
2,226,936
A IA economiza tokens com a minha linguagem de programação. Veja esse benchmark que eu fiz pro meu blog baseado em regras que eu defini para minha linguagem se sair melhor 🤘
212