MaximillianNYC Profile Banner
Maximillian Piras Profile
Maximillian Piras

@MaximillianNYC

Followers
1K
Following
24K
Media
1K
Statuses
6K

AI UX @yutori_ai | prev @headlinerapp_ @8tracks

WWW
Joined February 2010
Don't wanna be here? Send us removal request.
@MaximillianNYC
Maximillian Piras
1 month
As a user, I want a website that learns to use other websites so that I can use fewer websites.
0
0
6
@MaximillianNYC
Maximillian Piras
1 day
Agent Skills are cognitive artifacts* for LLMs ;) *see Don Norman’s 1991 paper
@robertnishihara
Robert Nishihara
2 days
The thing that is exciting *to me* about @AnthropicAI's "agent skills" announcement is that it provides a step toward continual learning. - Rather than continuously updating model weights, agents interacting with the world can continuously add new skills. - Compute spent on
0
0
0
@MaximillianNYC
Maximillian Piras
1 day
Good ready by @zehf on the risks of UI decisions growing or dulling user agency: “Often, we don't want to be better navigators, or our use of math is so trivial it doesn't justify constant practice. And that's fine. So long as it's a deliberate choice.” https://t.co/UAkaBilQwv
Tweet card summary image
zehfernandes.com
1
1
3
@MaximillianNYC
Maximillian Piras
2 days
My xp in agent builders thus far aligns w/ this reasoning: "A probabilistic-first system actually allows you to skip a lot of the complexity inherent in a visual builder. You don’t need to worry about all of the branching logic, or exceptions, or looping back."
@Padday
Paul Adams
7 days
When we look back, one of the defining product challenges of this time will be how we blended determinism and probabilism into a single product experience. People need both. Early last week, @OpenAI announced a deterministic heavy, workflow oriented Agent builder. A few days
0
0
3
@deviparikh
Devi Parikh
2 days
Our Chief Scientist, @DhruvBatra_, on proactive vs reactive agents, long-horizon RL with Scouts running for weeks, how agentic traffic changes the attention economy of the web, and why reimagining humanity's interface with the web requires generative UIs + intelligence built
@StackOverflow
Stack Overflow
2 days
🎙️ Dhruv Batra, co-founder of @yutori_ai, sits down with us to explore the future of AI agents, how AI is changing the way people interact with the web, and the challenges that proactive AI agents may face. https://t.co/gPy4Iiy2Km
2
5
50
@MaximillianNYC
Maximillian Piras
3 days
Prototype of an input field that reveals additional settings as you type 🙈
0
0
5
@MaximillianNYC
Maximillian Piras
3 days
Sora is the new GarageBand, this is such an apt take from @benthompson
1
1
1
@MaximillianNYC
Maximillian Piras
4 days
While playing with Claude Code, I was really loving its loading animation. So I asked it to make me a CLI Loader WYSWYG tool :)
0
0
4
@alessio_joseph
Joseph Alessio
5 days
RIP drew struzan, the illustrator behind some of the most iconic film posters of all time
1
1
15
@MaximillianNYC
Maximillian Piras
6 days
Dog rescue is by far my fave Scouts success story thus far 🐶
@abhshkdz
Abhishek Das
6 days
We've been quietly shipping improvements to our agent infrastructure, and it's getting really good.
0
1
2
@radbackwards
dar
6 days
It’s 2027. There’s no more Ubers, just robotaxi’s…. Some of us still pay in cash to ride in 2002 BMW Taxi’s up University Ave— similar to how they ride horse carriages thru snowy Chicago streets— to remind us of the good old days.
11
4
188
@MaximillianNYC
Maximillian Piras
6 days
Similarly, friends should never 100% settle ‘debts’ with each other. Loose accounting means you will forever remain on each other’s books & the numbers will work themselves out over the course of your shared lifetimes.
@melodyskim
melody kim
7 days
There are no healthy, convenient relationships. Any semblance of one is an illusion. Strings always remain attached.
0
0
1
@MaximillianNYC
Maximillian Piras
7 days
Here's a peek at my new toy LLM app called Zoomer: a simple "semantic zoom" interface to easily distill ⇔ extrapolate text as you write. It's just a fun thought experiment designed in code, but LMK if you wanna play with it & I'll deploy / open source soon!
@MaximillianNYC
Maximillian Piras
14 days
New smol wknd project in the works.
2
0
8
@skirano
Pietro Schirano
7 days
The next billion dollar agents will be what I call “proactive agents.” These are the ones capable of predicting user intent before a request. No one has figured out a reliable way to do this just yet. We are close.
212
80
1K
@MaximillianNYC
Maximillian Piras
8 days
Another @stripepress addition into my antilibrary. Excited to read this one by @dwarkesh_sp & @g_leech_ ⚖️
0
0
3
@MaximillianNYC
Maximillian Piras
10 days
So much fun to see @SANDYalexg live! Such a great Scout find… will ask it to get closer to the stage next show tho🤘
@MaximillianNYC
Maximillian Piras
21 days
Scouts finally found me a deal on Alex G tickets, can't wait to see him live!
0
1
2
@MaximillianNYC
Maximillian Piras
11 days
Played around w/ some @areatechnology_ inspired aesthetics, but wound up looking like a cheap version of liquid glass... still *thinking*.
@MaximillianNYC
Maximillian Piras
14 days
New smol wknd project in the works.
0
0
1
@MaximillianNYC
Maximillian Piras
12 days
0 to 1 is so fun :) it's always endearing to think back to when the thing you helped build was an earnest napkin sketch
0
0
4
@MaximillianNYC
Maximillian Piras
13 days
Hope you have a happy It's So Over / We're So Back Day to all who celebrate 🎂
@vasumanmoza
vas
13 days
OpenAI just murdered half of SF's startups. Was an impressive DevDay today, here's everything you missed (bookmark this): 1. Apps SDK: Shopify App Ecosystem / Whop App Ecosystem but for ChatGPT. Presumably (?) with monetization down the line. Build apps that anyone can use,
0
0
2
@krispuckett
Kris Puckett
13 days
Clever and nostalgic @OpenAI
1
1
5