ArmanAssadi Profile Banner
Arman Assadi Profile
Arman Assadi

@ArmanAssadi

Followers
4K
Following
8K
Statuses
5K

Co-Founder & CEO of https://t.co/PPeIWw7BVq | Co-Host of @alfalfapod | we are gods with anuses

America’s Finest City
Joined February 2009
Don't wanna be here? Send us removal request.
@ArmanAssadi
Arman Assadi
10 months
Steve Jobs wrote himself this email as a reminder. Fascinating to see the parallels between it and this quote below: “A hundred times every day I remind myself that my inner and outer life is based on the labors of other men, living and dead, and that I must exert myself in order to give in the same measure as I have received and am still receiving.” - Albert Einstein One of my all time favorite quotes. We’ve lost hold of this fundamental truth today in our achievement-junkie, independence-focused, “self-made” narcissism-celebrating culture. Let the dead speak.
Tweet media one
6
20
91
@ArmanAssadi
Arman Assadi
4 days
fwiw I asked a pretty complex, nuanced, multi-variable question about our pricing model to ChatGPT 4o, o3, o1, DeepSeek R1, and Claude 3.5 Sonnet. The best answers came in this order: 1. R1 2. Sonnet 3.5 3. o1 4. 4o 4. o3 (yes it was worse than 4o and o1) I think this says a lot of things that could be unpacked but the main thing here is pick the right model for the right job.
0
1
4
@ArmanAssadi
Arman Assadi
5 days
RT @ScottAdamsSays: Complexity is always a cover for fraud. In every domain.
0
3K
0
@ArmanAssadi
Arman Assadi
8 days
RT @growing_daniel: Luke being in charge of our money is an incredible shift. I didn’t know this was an option, amazing development https:/…
0
808
0
@ArmanAssadi
Arman Assadi
9 days
This is amazing
@karpathy
Andrej Karpathy
10 days
There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It's possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good. Also I just talk to Composer with SuperWhisper so I barely even touch the keyboard. I ask for the dumbest things like "decrease the padding on the sidebar by half" because I'm too lazy to find it. I "Accept All" always, I don't read the diffs anymore. When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I'd have to really read through it for a while. Sometimes the LLMs can't fix a bug so I just work around it or ask for random changes until it goes away. It's not too bad for throwaway weekend projects, but still quite amusing. I'm building a project or webapp, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.
0
0
0
@ArmanAssadi
Arman Assadi
10 days
@mrymz_2 Couldn’t agree more. I think this is my new favorite topic.
0
0
2
@ArmanAssadi
Arman Assadi
11 days
What do you think about working weekends? Overkill or superpower?
@elonmusk
Elon Musk
11 days
Very few in the bureaucracy actually work the weekend, so it’s like the opposing team just leaves the field for 2 days! Working the weekend is a superpower 😂
0
0
2
@ArmanAssadi
Arman Assadi
11 days
RT @SERobinsonJr: Starship. Human for scale.
Tweet media one
0
2K
0
@ArmanAssadi
Arman Assadi
12 days
Excellent write-up — this investor clearly understands the space.
@omooretweets
Olivia Moore
14 days
🚨 New @a16z thesis - an update on AI voice agents! 2024 was a huge year for voice agents, with massive model breakthroughs + hundreds of new startups. How has the space evolved, and where do we see opportunity in 2025? Our update + market maps for B2B and B2C 👇
Tweet media one
0
0
1
@ArmanAssadi
Arman Assadi
13 days
RT @balajis: It might turn out that the GPT wrapper has more of a moat than GPT.
0
501
0
@ArmanAssadi
Arman Assadi
16 days
RT @signulll: some people didn’t get my apple post about execs smiling this morning, so let me break it down a bit. the r1 news is massive…
0
808
0
@ArmanAssadi
Arman Assadi
17 days
RT @morganb: 🧵 Finally had a chance to dig into DeepSeek’s r1… Let me break down why DeepSeek's AI innovations are blowing people's minds…
0
13K
0
@ArmanAssadi
Arman Assadi
17 days
0
0
1
@ArmanAssadi
Arman Assadi
18 days
Answer the call and begin the journey
@RobertGreene
Robert Greene
18 days
Most people wait too long to go into action, generally out of fear. They want more money or better circumstances. You must go the opposite direction and move before you think you are ready. It is as if you are making it a little more difficult for yourself, deliberately creating obstacles in your path. But it is a law of power that your energy will always rise to the appropriate level. When you feel you must work harder to get to your goal because you are not quite prepared, you are more alert and inventive. This venture has to succeed so it will.
0
0
2
@ArmanAssadi
Arman Assadi
18 days
RT @venturetwins: The LLM world moves fast 😅
Tweet media one
0
1K
0
@ArmanAssadi
Arman Assadi
18 days
For me the most exciting thing about Operator is that it will be a step toward bringing #7 to life
@dharmesh
dharmesh
18 days
Have been playing with OpenAI's "Operator" agent all weekend. Some early thoughts: 1) The fact that it's already hosted and you can just start using it means it'll be more widely used than Claude's "Computer Use" feature. 2) Operator will be able to make websites that haven't had APIs available (for whatever reason) behave as if they did. They'll now be automatable. 3) Operator is a bit like tool calling in LLMs. Except it uses the entire Internet as its set of "tools" (whether the site has an API or not). 4) We might see a form of caching/pre-compiling for widely used functions. Essentially, Operator could write its *own* API for things that are commonly needed, and then just use that to get that particular task done. 5) Operator can use itself behind the scenes to distribute a long list of tasks across multiple instances of a virtual browser. 6) Operator will likely be able to reason through how a task could be parallelized. Example: Task involves paging through 50 web pages of results. Instead, it could say: Operator A will start at 1st page and go forward. Operator B will start at last page and go backwards. They'll "meet in the middle" when all pages are done. 7) Current websites were (mostly) built for humans and APIs were (mostly) built for developers (a special type of human). Mobile sites were built for humans using mobile devices. I wonder if we'll see websites created specifically for use by agent AIs. Anyone else enamored with Operator yet? What kinds of use cases are you trying it out on?
0
0
0
@ArmanAssadi
Arman Assadi
18 days
Friendly reminder to never waste time on mediocre books that don’t slap
Tweet media one
0
0
2
@ArmanAssadi
Arman Assadi
18 days
RT @naval: Turns out that instead of scraping the web to train an AI, you can just scrape the AI that scraped the web.
0
2K
0
@ArmanAssadi
Arman Assadi
19 days
Jordan Peterson <> Tony Robbins 🔥
0
0
1