Renato Valdés-Olmos retweeted
once teams have figured out building w AI (ai powered eng, prototypes, etc.) their bottlenecks move upstream and downstream
upstream issues:
- pm capacity
- no good ideas
- rigid org structures + process
downstream issues:
- pr review / capacity
- verification
- commercial (slow gtm, no $$ value)
how to fix:
- redesign your org for new physics
- focus on ambitious ideas, not just features
- make eng own more requirements so PMs can talk to customer and cultivate big swings
- automate your PR review + risk scoring
- cloud/preview environments + tooled agents for verification
- tighter PM <> GTM partnership
all solvable, but takes a leader willing to force the changes necessary
Renato Valdés-Olmos retweeted
i love the stylistic choices in into the spider-verse, where miles morales is animated at 12fps while everything around him runs at 24fps. that and the juxtaposition of the square enix hd-2d style made me want to see what this could look like in UI
the only way i could make it work was by over-exaggerating the movement, which kept it feeling intentional. subtle movement at a low frame rate just felt like a performance issue
For everyone looking to switch away from Slack, give @getbureau a try. It was built in about the same time it took them to ship dem nifty 🎃 statuses.
Renato Valdés-Olmos retweeted
The benefits of abundant intelligence will accrue far outside of San Fransisco.
Spent the last year writing about capital, sovereignty, and the coming reordering of economic power.
THE COMMANDING HEIGHTS
Commanding-Heights.com
Renato Valdés-Olmos retweeted
I've lost count of how many times I've reread Designing the Star User Interface (1982) from Xerox PARC. Every time, I'm struck by how sincere and thoughtful the writing is. Really want to share this with anyone thinking about the next generation of UIs.
A few things that particularly resonate with me (there's so much more in the article):
1. A consistent conceptual model matters. I believe the next generation of UIs should be generative and adaptive, but still follow a coherent, learnable set of 'physics'. Malleability shouldn't come at the cost of predictability.
2. Simplicity should be a first-order design principle (something LLMs are bad at). Fewer buttons, clicks, or keystrokes don't necessarily make an interface simpler. Nor does adding more. "simple things should be simple; complex things should be possible"
3. Plan to throw one away; you will, anyhow. Many prototypes are meant to challenge assumptions - they should be designed to be thrown away, rather than promising to deliver the throwaway to users.
** What stays with me most is just how much nuanced, deliberate thinking went into the WIMP interfaces we've inherited, and are now eager to reinvent.
We need at least the same depth of thought and care. Still have a long way to go.
Renato Valdés-Olmos retweeted
The coding interview is dead, at least for my team at Anthropic. I've seen enough to know that it no longer provides signal. I removed it from loops this month.
I have a spicy, optimistic take - I think there's an opportunity for junior engineers: felixrieseberg.com/hunger-be…
Renato Valdés-Olmos retweeted
Brutal truth from @Lethain
“If you're in [middle] management (…) you sort of have to decide, am I retiring now or am I re-skilling?”
He did just this: step down from leading a large team (250 ppl) to be hands-on with a small team (25 ppl), ship code & re-learn the new “rules”
Why a 3x CTO went back to a 25-person team. From Will Larson @lethain, Imprint CTO and 4x book author.
"It's a hard time in middle management, right? There's a lot more competition for these roles. What I found personally very rewarding is getting back into the details of writing software.
When I left Carta, my team was around 250. When I joined Imprint, the team I was leading was 25.
This a unique moment to actually re-skill. If you're in management, I think right now either you have to get back into the details, or the world is changing so radically that the profession you know will not be valued in three years. You sort of have to decide, am I retiring now or am I re-skilling?
I think there has to be some openness to taking a backward step or a lateral step. What I knew I wanted was getting back into the details and relearning how to do software in the current age. That's what I believe is going to be the unlock to the longevity of my career."
Renato Valdés-Olmos retweeted
we've gotten so obsessed with speed that taking time to think now feels like a bug. it's a feature.
Renato Valdés-Olmos retweeted
Lesser known academic theory about horror/sci-fi movies is that they’re often used to externalise society's deepest traumas into monsters and dystopias.
So I used Argon to tap into cultural symptomatology and psychoanalytical film theory to create this website which traces 100+ horror/sci-fi works and the social themes that they actually mirror across 7 eras.
Enjoy your days upto the Halloween with this project!
culturalcanary.ai.studio
Used Gemini for deep research, Argon in Antigravity used Nano Banana to generate the movie figurines + create the build, AI Studio to add the final finishing touches.
pepsi-arnell-021109.pdf
never forget
rebrandlog.com/rebrand/pepsi…
I stopped giving design candidates homework in 2020. Six years later, I’m still happy with that decision.
Some thoughts on what to do instead, and its limitations.
renatovaldes.com/essays/agai…
Renato Valdés-Olmos retweeted
Your team. In good company.
Bureau is now in beta. Shared conversations, project knowledge and AI agents, together in one workspace.
Request access: getbureau.app
today's light read
interpretability might be the most important scientific problem in the world right now, as I don't think we can solve alignment if we don't know what we're building. I think I have good news about it. I co-authored early work in mechanistic interpretability at OpenAI and came away pessimistic, because neural networks appeared to fold into arbitrary tangled knots. years later, I started openai's first large-scale automated interpretability project, finding about 2% of neurons legible with the primitive tools we had
with modern agentic tools and a much closer look, the picture looks totally different. the vast majority in models up to trillions of parameters are legible, and their circuits beautiful. many famous examples of tangled neurons turned out to be misunderstood. The roadblocks I expected mostly aren't there. I now think ambitious interpretability will be solved by the time we need it most, and possibly quite soon, as automated scientists keep improving. it may be a bitter lesson for interpretability: the simple thing, at absurd scale, works
already-clean.goodfire.com/
On Friday I started building my own harnass, today it has a functional RSI loop, a MacOS client and iOS app. Called it Bureau.
this has really gotten out of hand nitter.cf/renn/status/2110009417…
Introducing @getbureau: a shared workspace for creative teams and their AI agents. Now in invite-only beta. getbureau.app