Reed Oliver

[Reed Oliver @ReedOliver11]

https://t.co/8cmgH025cm

[Reed Oliver @ReedOliver11]

Too many people having one night stands with claude to churn out slop posts for clout. You gotta be awkwardly rubbing your wiener on the details like it's an eighth grade mixer. I gotta see signs that you tried homie

[Reed Oliver @ReedOliver11]

@nrehiew_ QB stuff tripped me up too. I think the explanation in the appendix is good but I was too dumb to get it initially. This is where I ended up

[Reed Oliver @ReedOliver11]

@gordic_aleksa Great writeup. HTSYM sent me down the same rabbithole and I ended up tracing collectives from XLA and rebuilt things in Pallas. Fun to validate profiler numbers against what's on the spec sheet https://t.co/DGWXgJdley

[Reed Oliver @ReedOliver11]

@PatrickToulme Not directly the point you’re making, but this reminds me of something Reiner said wrt MatX, ie that if your biggest customers plan to write kernels at lowest levels of abstraction as they can anyway, just make that the default

[Reed Oliver @ReedOliver11]

@A_K_Nain Righteous

[Reed Oliver @ReedOliver11]

@thaiscbranco_ Just giving y'all a hard time. Best of luck!

[Reed Oliver @ReedOliver11]

We asked 100 SF founders to name a startup

[Thais Castello Branco @thaiscbranco_]

We’re excited to introduce Taste Labs.

Our mission is to end AI slop. We’re building the data and infrastructure layer to give AI models and agents taste.

And today we’re coming out of stealth, announcing our $18.5M seed funding, co-led by @CRV and @AmplifyPartners

AI has nailed objective domains and made it easy to generate anything. But it still feels off. Now, the challenge is judgement. What fits, what feels like you, what’s GREAT. This requires turning a fuzzy, subjective domain into something we can measure and codify. We’re starting with design.

There are two sides to cracking this, the foundation model layer and the agent layer:

- We’ve already been working with the top frontier labs to evaluate and improve their models, crafting the right post-training data and RL environments.
- We’ve also been working with app-layer companies to build the context and verification tools for their agents to produce better, more on-brand, more creative outputs.

We want a future where AI feels right.

If you’re passionate about this mission, join us!

[Reed Oliver @ReedOliver11]

Toucan Sam was really on to something with this

[Reed Oliver @ReedOliver11]

Goals are for suckers. I should’ve just let myself be

[47fucb4r8curb4fc8f8r4bfic8r @47fucb4r8c69323]

I have been struggling with this ever since I acquired fuck you money.

The issue for me is that I tried to do many things and gave up on them the first time I had a big win. This is a personality trait of mine, and I’ve come to live with it: I am a sore winner. As soon as I win one stage of a game, I immediately lose interest in the game itself and start to think about how I can beat the game. Remember, winning and beating a game are very different.

As a result, the things I tried to do were never fully done. I never became a novelist, I left academia a couple of years into my first tenure-track job, I abandoned journalism because it didn’t pay enough, and then when I finally found something that I could do for a long period of time—sell-side research—I tried to maximizing the pay and minimizing the hours I worked at that job instead of trying to go to the next stage of the game, which would be to go buyside (and then move up to PM and then principal and then become a Dan Loeb type).

Then when I had enough money that I didn’t need to keep working, I looked back and realized that I didn’t become a novelist because I’m not that guy. I didn’t become a professor because I’m not that guy. I didn’t become a journalist because I'm not that guy. I didn’t become a hedge fund tycoon because I’m not that guy.

But the guy that I am emerged throughout all of it: I am a cool island guy. I’m borrowing that phrase from a great philosopher, but it means what it says: I’m just a guy, and I’m a guy who needs to be on an island of some sort, and only then can I be cool.

And it’s really quite horrifying to realize you’ve spent your entire life escaping being that guy when you thought you were trying to be that other guy. And as a result, you became that other other guy which was never your target.

The mistake was always having a target. Goals are for suckers. I should’ve just let myself be.

But I cannot tell you this, because maybe you have a different personality to me. Maybe you’re less desperate than I was. Maybe you weren’t threatened as much. Or maybe you just weren’t as scared as I was, so you can just lazily stay at home and goon and not be scared of loneliness, or you can just lazily stay at home and not work and not be scared of destitution, or you can just lazily stay at home and not study and not be scared of being stupid.

What drove me was always not being that guy. It was never becoming that guy.

Negative hagiography is my religion.

I don’t want to go to Heaven—I want to escape Hell.

And I did, and that should be enough. It really should. I want it to be enough. I pray that it be enough.

But it’s not enough. For now.

Da mihi mortalitatem et continentiam, sed noli modo

[Mezz @for_yield]

@BoringBiz_ likes ball but doesn't love to hoop

[Reed Oliver @ReedOliver11]

https://t.co/yfzjwfMPcg

[Boze Herrington, Library Owl 😴🧙‍♀️ @SketchesbyBoze]

The great sin of our age is incuriosity. People are not motivated to learn, they are not motivated to pursue knowledge. We need a generation of students that are hyper-literate, a generation that rejoices in reading. If there's any hope for the future, it begins in the library.

[Reed Oliver @ReedOliver11]

@zetalyrae There is a phenomenology of writing that is lost on those who have never tried in earnest

[Reed Oliver @ReedOliver11]

Right ahead of this one

https://t.co/95wLF3IaYt

[eigen moomin @eigen_moomin]

all the startup founders are hot now. jacked with abs. gigawatt smiles. beaming supreme confidence and ineffability. i do not trust these djinn.

what happened to the troglodytes? spines bent, skills bartered with scoliosis from years of sitting like lobsters programming Altairs?

[Reed Oliver @ReedOliver11]

I think about this tweet more than any other tweet

[Rota 🚪🧎‍♂️ @pli_cachete]

We’re not that different from bears you know.

[Reed Oliver @ReedOliver11]

Ball is ball

[bubble boi @bubbleboi]

Pod bros message me about astera all the time asking me why I’m bullish.

It has nothing to do with copper, optics, retimers nada. That is their biggest business today so it makes sense why people anchor on that but I see what’s coming next.

The engineering problem Astera is solving with CXL & the stuff around it is extremely valuable. People think CUDA is a moat, lol not even close and you see it with ASIC-X. The real moat is the next step & Nvidia is losing badly there.

[Reed Oliver @ReedOliver11]

Reeds Day 3 at mlsys:

[Reed Oliver @ReedOliver11]

Reeds Day 1 at mlsys: Where are the TPUs

[Reed Oliver @ReedOliver11]

Rolling in to Bellevue like this tomorrow morning

[BuccoCapital Bloke @buccocapital]

I can’t stop thinking about this post. If you do one thing today, I encourage you to give it a thoughtful, thorough read…

And then commit to never living your life this way. Life has wasted success on the people described in this post.

It really is completely pathetic. They say that comparison is the thief of joy - look no further than this post for validation it is indeed true.

On their deathbed they will realize they have lived their life completely wrong. Don’t let it be you.

[Reed Oliver @ReedOliver11]

@DeepDishEnjoyer Might have to dust off Safe As Milk now