@foundrarav
Account based inUnited States!
About this account
- Account based in
- United States
- Connected via
- Web
X says this location may be affected by a proxy or VPN.
Account-level information from X, not a live location or the device used for a specific post.
From ideas to investments 💡 | Ex-founder journey 🐾
New York City
Joined June 2023
- Tweets1K
- Following292
- Followers30
- Likes5.1K
AravBizBuilder retweeted
From the Colossus interview with Vélez: "Every major industry around the world has been reinvented by technology companies except banking. Ninety-five percent of the world's financial services still sit with large incumbents. It's almost like saying 95% of the world's video delivery is still owned by Blockbuster, or JCPenney instead of Amazon, or taxis instead of Uber." $NU $AMZN $UBER
AravBizBuilder retweeted
$CBRS is probably good value here sub $180.
CS-4 just doubled wafer performance and said inference can run up to 30x faster than GPUs.
Here's why this matters:
Speed is where lots of demand is heading. Coding agents make hundreds of model calls in a loop so faster decode -> a job finishing quicker.
GPT 5.6 for example will have a premium fast mode which suits customers like:
-> Quant funds: For them time is money.
-> Pharma and research agents: $GSK has talked about intelligent research agents on $CBRS.
-> Voice and real time support
$CBRS unlocks a new level of speed, and in AI...speed is what people are happy to pay a premium for.
If you use these for work, you will have loads of use cases:
- Analysis & BI becomes proactive, not reactive
- Optimise your paid marketing spend
- Catch and address product issues proactively
- Never manually update a CRM again
AravBizBuilder retweeted
agents can already find what you want to buy but paying is still where it stops.
so i was researching some days ago and randomly came across @agentcardhq.
apparently they're backed by @ycombinator and the idea is simple.
give the agent a card.
say an agent needs to order something online, book a ride, or pay for food. it can get there, however, checkout still needs you.
Ideally, you don't want to hand over your real card, and a lot of sites still block these agents.
that's where they come in
you can save the card you already use and approve the payment on your phone or they make a new card, with a spend limit.
it can be one-time or reusable.
also, the money sits in usdc on base, so it spends from a limit, not your whole balance.
they're still pretty early and mostly useful if you're building something that buys for users.
you can check for more details on their website in the comments.
disclaimer: this isn't a paid post.
I just like what they're building and felt like sharing.
AravBizBuilder retweeted
Steve Eisman owns Nvidia stock. He says five customers account for 70% of Nvidia's receivables, and calls it a warning flag.
"And I own the stock, okay?"
"Revenue growth of over 100%, like that's insane. So that would say the AI story is great, until you read the 10-Q, which came out that night."
"Note 7 says that 70% of Nvidia's accounts receivable as of the end of July came from five customers."
"Warning flag. Not the end of the world. Warning flag."
- Steve Eisman, investor and Nvidia shareholder, interviewed by New Money. Video uploaded October 2, 2026. Nvidia's July 26, 2026 Form 10-Q reports five direct customers at 22%, 14%, 13%, 11% and 10% of receivables. This is a receivables snapshot, not their share of revenue or a prediction of default.
AravBizBuilder retweeted
Her startup was about to receive Rs 2 crore in funding from Singapore.
The only condition? She had to register the company in Singapore and transfer the patent to the Singapore government.
She declined the offer and decided to build in India instead.
The story of @Smithadevigere, founder of Devigere Biosolutions, who is building India's first fully water-soluble plant-based protein, is winning hearts and minds on @mundhebanni.
Catch the full episode with English subtitles here: 👇
youtube.com/watch?v=7qAMKptV…
AravBizBuilder retweeted
Patrick Collison says one Stripe engineer merged more than 600 PRs written with AI in H1. One was reverted.
"I was speaking with one engineer at Stripe this morning, and over the course of H1, he had more than 600 pull requests merged. Every single one was written with AI."
"And exactly one of those pull requests had to be reverted, which at least suggests that it is possible to have this enormously accelerated development rate with still empirically quite high reliability."
- Patrick Collison, Stripe co-founder and CEO, speaking on the official Claude channel.
hello buddy, how goes it
This was almost way too easy, @natfriedman @alexandr_wang 😮
I turned my Xteink eReader into an always-on Muse pocket display for Calliope (my Muse) to display itself and show status updates.
And this attaches with MagSafe to my phone, so I can always see relevant updates from Muse by just turning my phone around.
The setup flow in the Muse app was super smooth and you can just ask Muse to load or automate content on the display.
I genuinely love the hacker-friendly aspect of this. Well done.
Calling classifiers "decision models" is a crime against machine learning. Those models decide *less* than LLMs, as you can think the CoT as a form of serial processing of information that such classifiers lack. Those models can be either *small* LLMs that are not let think, but instead the logits at the last position of the prefill are used to categorize in classes or, when they are BERT-alike, they compress the input meaning and project a class: in both cases they decide a lot less than an LLM.
AravBizBuilder retweeted
After nearly six incredible years building spacecraft, I’m beginning a new chapter as an entrepreneur in residence (EIR) at @paradigm.
Varda was a grand adventure and a one of one experience I’ll cherish forever. I had the rare privilege of joining the founding team and helping take a wild idea from zero to one, from a blank sheet of paper to hypersonic vehicles coming home from orbit. I’m incredibly proud of what we built together. Thank you to Will, Delian, and the whole team.
Paradigm is giving me something incredibly cool. I’ll have the freedom to search for the next important thing worth building. This might look like a radical departure, but I see it as a chance to step back, look broadly, and follow my curiosity. An EIR sits somewhere between researcher, inventor, and founder in waiting. Paradigm feels like a natural home for that kind of exploration, full of earnest nerds who pair deep technical curiosity with research.
My role is to explore important problem spaces, develop conviction, run experiments, and eventually turn the most promising direction into something real. In many ways, it feels like a natural extension of my hobbies and passion projects.
I’ll be in SF during the week and LA on weekends. I want to use that time to meet as many interesting people as I can. Let’s get coffee and talk about energy, compute, artificial intelligence, and manufacturing. I want to hear about your most ambitious ideas and deepest technical rabbit holes.
AravBizBuilder retweeted
I'm biased (I coach PMs for a living) but Andrew Chen is right. And honestly I think he's underselling it.
His whole argument is one line. When anyone can build, the person who decides WHAT to build becomes the bottleneck.
After coaching hundreds of PMs through this shift, here's who actually becomes that bottleneck.
✦ PMs who build the prototype first and write the spec after will be in crazy demand.
✦ PMs who can look at 15 working prototypes and pick the 3 worth shipping will be in crazy demand.
✦ PMs who write evals will be in crazy demand. Judging AI output is becoming what judging designs used to be.
✦ PMs who treat "technical enough" as table stakes will be in crazy demand. At the frontier labs it's already in the job description.
✦ PMs who'll own whatever the team is missing that week (positioning, pricing, a data pull, onboarding copy) will be in crazy demand.
✦ PMs who get that building is basically free now, so the whole job collapsed into choosing, will be in crazy demand.
And look, this isn't a prediction. OpenAI already pays its median product manager $860K. Job postings at OpenAI, Anthropic, and Google DeepMind already ask PMs to write evals and prototype with code.
Meanwhile the tickets-and-standups version of the job is getting automated by the same tools making this version valuable.
PS if you want to learn all of this, I've written a deep dive on each piece (a free trial unlocks them all):
1. Become an AI PM: news.aakashg.com/p/become-an…
2. Claude Code for PMs: news.aakashg.com/p/carl-vell…
3. AI prototyping guide: news.aakashg.com/p/ai-protot…
4. How to write evals: news.aakashg.com/p/ai-evals?…
Everyone can build now. Being the bottleneck just became the best job in tech.
AravBizBuilder retweeted
I showed Zapier's CEO three versions of the same PM work and had him grade each one live against Zapier's AI fluency rubric.
A PRD written with ChatGPT: Capable.
He named what was missing. No prototype, which he now expects because AI makes one cheap. And no evidence. Where did customers ask for this? Gong calls, Zendesk tickets, Reddit threads, posts on X.
His line: "If there are things I know that you don't know, that's probably not a great signal."
A working prototype plus a PRD that cited the evidence: Adoptive.
An operating model for a whole product team: Transformative. Skills hold the recurring work. Subagents review in parallel, including a skeptic that argues nothing should ship. The agent reads Zendesk, Gong, Amplitude and Linear directly, and a human gate sits before anything merges.
"This is not how product and software was built three years ago. Not even close."
The catch he added: about 1 PM on a team of 40 should hold that rating. The floor moved. What was transformative in 2024 is adoptive now.
The rubric, the three artifacts and his notes on each:
news.aakashg.com/p/zapier-ai…
ALT Zapier: the 4 primary components of AI fluency. Mindset, strategy, building, accountability.
AravBizBuilder retweeted
People keep responding to my posts about Dot saying, “can’t you just do that in Codex”
And the answer is yes, sure, but in an entirely different and way more complex way.
Jason illustrates the difference perfectly here:
to be clear you can do almost everything in dot/space with vanilla codex
for example before those tools i just had
* 15 pinned tasks each with a heart beat automation
* Github repo for a memory vault based with agents.md files everywhere so all your agents can contribute
* make sure both your technical and non technically team know about Github and can clone the repo and contribute based on my rules
I just had 4-5 different mini apps served to render artifacts
and then do a bunch of 1:1s to set up the rest of your technical and non technical teams to use this workflow to collaborate and make sure their automations are set up on their computers
or you can be like 'yo dot whats new, lmk'
AravBizBuilder retweeted
🚨 OpenAI is reportedly targeting $30B in fresh funding at a $1.4T valuation.
And with Sam Altman saying OpenAI won’t IPO in 2026, the public listing could be pushed to 2027.
JOIN FREE VIP GROUP :
t.me/wiseadvicetrade_bot
AravBizBuilder retweeted
Opus 5.5 is making us rewrite our entire harness because of how it already does so many things we had built in our harness.
Usually this always happens with a new model launch. For example, when 5.6 came, a lot of the test driven framework that was enabled in the harness was kind of redundant.
But the amount of change that 5.5 has brought is truly insane. 4.6 levels
