Ayush Yadav

Product, writing occasionally & dreaming mostly

[Ayush Yadav @psuedosapien]

Let the game begin

[Ayush Yadav @psuedosapien]

@jpatel41 This is the kind of hiring post that makes me want to apply before even seeing the JD

Hospitality → ecommerce/payments → AI × investing as a PM so far.

Mostly just learning unfamiliar things fast, owning messy problems, and obsessing over the details until they work

[Ayush Yadav @psuedosapien]

@lennysan Throw me into an unfamiliar domain and give me little time, I’ll figure out how it works

So far: hospitality → ecommerce/payments → AI × investing as a PM

Looking for ambitious remote team <3

[Ayush Yadav @psuedosapien]

Looking for a few retail investors 👀

If you invest in stocks/MFs and bought (or seriously considered) an individual stock in the last year, I’d love to chat.

45min user research call + early product screens.

Coffee/Amazon voucher as token of appreciation🎁

DM “Interested” 🙌

[Ayush Yadav @psuedosapien]

Been pondering over this recently,
Narrowed down to a few insights-
1. In india under 1% of total land is being used for urban residential use>The infra including roads, schools, hospitals, etc. are highly concerntrated, also jobs>Large population fighting over small piece of pie

[Lt Col Sushil Singh Sheoran, Veteran @sushilskinner]

I am neither a communist nor an ultra socialist, but an apartment in Gurugram, which has third-world civic amenities, is being sold for Rs 271 crore. There is something extremely wrong with India. It is surprising that where you are taxed for an income of more than 12 lakh, starting at 4 lakh. 80 crore people are given free food. As soon as you earn 50 lakh, the tax slab goes up disproportionately, and the excessive cess takes charge. The income distribution is very uneven, and worst of all, there seems to be a particular class that gets special treatment—middle class, taxed as if it recalls French taxation before the revolution. There is something fundamentally wrong with India's economy. I was astounded to see technical professionals living in PGs in Gurugram or Bengaluru, enduring inhumane conditions because renting a flat is beyond their capacity, earning up to 1.5 lakh. There is something wrong. I wish someone could explain it to me.

[jo johnson @josbjohnson]

you must believe you are special and then go so hard, for so long, with such violent refusal to accept any other ending, that reality itself starts running out of ways to tell you no. you must wage a war daily against the ordinary outcome, until the belief you invented out of nothing in a room by yourself has been hammered into the world so many times that it stops being a claim and becomes reality.

[Ayush Yadav @psuedosapien]

Start with user experience and work backwards to the features, then the product and at last the technology

[Ayush Yadav @psuedosapien]

@zach_yadegari Some features deserve to be products- Users

[Ayush Yadav @psuedosapien]

@KESCoHQ @UPPCLLKO @UPPCLLKO @KESCoHQ

[Ayush Yadav @psuedosapien]

@aryanistweeting @miless_15 Splitting happens usually later; no one wants to split the bill while ordering food. (also, mostly college students split food bill)

[Ayush Yadav @psuedosapien]

@hrdktwt to the moon🚀

[Jon Yongfook @yongfook]

It takes about 20 seconds of convo to gauge if the passion is there in a founder who has an early stage product. The passion / curiosity about the problem space is what you need to grind through the early stages, find PMF etc.

If all you see is a deadness in their eyes because it's the 19th vibe-coded app they have deployed to production this year with no users, you can just tell they are NGMI.

Life is short.

Work on things you are genuinely interested in.

[Ayush Yadav @psuedosapien]

what did you build this week?

[Ayush Yadav @psuedosapien]

@hrdktwt curious, how are you scrapping job postings?

[vas @vasuman]

Hey look, they wrote an article about Varick Agents.

The pitch to companies is simple: you know you need AI, but it's actually very hard to do properly.

This is not a traditional software engineering problem. Balancing context, governance, access, and autonomy is a tricky problem that requires AI engineers you can trust.

It's also not a problem that SaaS off the shelf can help you with. There's a reason why there's been effectively 0 ROI from enterprise adoption of 'AI SaaS'. You just spent the last 3 years migrating to your NetSuite, and the AI that your company adopts should live on top of it, not force you to replace it. Same with your Ramp, your Salesforce, etc.

Let Varick come into your company, understand what your company does, and build AI agents that live on TOP of your systems of record, without forcing you to rip it out.

Vibe coding one workflow after the next doesn't work. There's a reason why our clients see 30-50% efficiency gains by month 3.

If you're C-Suite at a company doing over 250M ARR, schedule a call, I'll show you how to implement AI the right way.

Also we're hiring, aggressively. We can't keep up with demand. DM if you're a top 1% engineer - in-person SF only.

[Prayank Swaroop @prayanks]

You just raised $5M to build someone else's moat.

Here's what I mean.

Most AI startups I meet are competing on the wrong axis. They're obsessing over model choice — Claude vs. GPT vs. Gemini. They're fine-tuning on domain data. They're building slick interfaces on top of state-of-the-art APIs.

None of that is a moat. All of it can be replicated in weeks.

The founders I'm most excited about are competing on a completely different dimension: time.

Every session a user spends inside a well-architected AI system is a deposit. The system learns their editing patterns, their risk tolerance, their preferences — implicitly, without being told. After six months of daily use, that system knows how you work in ways you couldn't fully articulate yourself.

That's not a product feature. That's a compounding asset.

The architectural decision that separates these two worlds is simpler than most founders think: stateful vs. stateless agents. A stateless agent resets after every session — all that signal, discarded. A long-running agent retains it, learns from it, gets harder to replace every single week.

The switching cost of a great stateless AI product is zero.

The switching cost of a great stateful one, after two years, is enormous — not because of contracts, but because leaving means starting over.

I've written a full framework on this — covering the four depths of personalisation, the three RL signals that drive compounding, and where the research frontier is heading.

Link in the comments.

One question for founders building in this space: are you designing for state accumulation from day one — or is that an afterthought?

[Ayush Yadav @psuedosapien]

@nateobrienn how do I volunteer?

[Ayush Yadav @psuedosapien]

Also speaking to a few investors 1:1 (15 mins). Happy to offer early access in return 🙌

[Ayush Yadav @psuedosapien]

We're trying to make watchlists actually useful for long-term investors 👀

Right now, most of them just show price… but don’t clearly show what changed in the business

We’re building something better and would love your input

It'll directly shape what we build 💛

Link ⬇️

[Ayush Yadav @psuedosapien]

https://t.co/wSEP9FZBYI

[Ayush Yadav @psuedosapien]

We're trying to make watchlists actually useful for long-term investors 👀

Right now, most of them just show price… but don’t clearly show what changed in the business

We’re building something better and would love your input

It'll directly shape what we build 💛

Link ⬇️