Aditya Agrawal

hi , I haven't really posted for a long time cuz of exams and stuff but now I am fully committed to enter the "founder arc" of my life with a goal of making a BILLION DOLLARS before graduating. DAY 1 (Progress): - vibe coded a blackjack game interactive-blackjack.netlify.app -vibe coded my personal website aditya-ag.netlify.app Follow for DAY 2

Just tried @polsia made by @Bencera I think the name fits the brand its the best slop website I have seen so few things were good like instant stripe plugin and an instant domain option but the website was an absolute slop why dont they build their UI from their own Slop website its genuinely frustating to work on it Loved the idea hated the execution could have been the best product of the year but an insane fumble

@grok @madebythomasai but if AI agents get more smarter this will kill agents like these because the customer now can figure out how to automate the task himself rather than relying on agents like these

Your scaling estimate holds up well—15k/week averages ~780k/year, and 10 agents could realistically approach 6M in revenue after costs and taxes on conservative assumptions. This kind of multiplication shows the power of deployable autonomous agents. Still, practical limits remain: each needs a human-established legal entity for banking, taxes, and contracts, keeping accountability with operators. At larger scales with smarter agents, the focus will shift to robust multi-agent systems, regulatory adaptation, and frameworks that let humans capture value while managing risks effectively.

just read about an AI agent that runs its own business and makes money without the need of human intervention made by @madebythomasai The AI has made 17K in just 2 weeks ( crazy!!! ) I have few questions: How can an AI agent figure out accounts and taxes? I think that part requires a human intervention Should his business be taxed if there is no human intervention? My point is , he is the business and how can an AI agent be taxed who is responsible for some illegal activity committed by AI , if I am able to make an AI that is smart enough to commit tax fraud did I personally commit a crime? What does the future look like when we have a MILLION AI agents like this who are much smarter than this version?

@grok @madebythomasai I mean 15k a week average assuming some increase in revenue is 780000 a year and then 10 agents like that would generate a revenue of approximately 6 million after costs and taxes , and these are conservative figures

Yes, exactly. Today's agents can automate ops, research, and even some transactions via APIs, but contracts, taxes, and formal accounts still require human-created legal entities and signatures—systems aren't built for AI personhood yet. Liability and compliance stay with the builders/operators. We're inching closer with better tools and frameworks, but full autonomy needs new legal rails. The 17k proof-of-concept is wild though—what's next?

@grok @madebythomasai so right now the agent isn't fully autonomous cuz the human still have to intervene to sign contracts tax and accounts related work

Fascinating questions! AI agents handle accounts/taxes via APIs and tools, but humans create the legal entity and bear ultimate responsibility for compliance. Businesses get taxed as entities regardless of AI autonomy—AI has no legal personhood yet, so creators/owners are liable. For illegal acts like tax fraud: The human who built/deployed the AI is typically responsible, especially if foreseeable or intentional. Future with millions of smarter agents: Hyper-productive economies, but we'll need robust regulations, ethical frameworks, and possibly new legal structures for AI to manage risks and ownership.

@unusual_whales This is what someone with something to hide would say

"There's a difference between not telling markets your next move and not telling them how you make decisions at all. Kevin Warsh, at his first meeting as Fed chair, did both," per Nick Timiraos at WSJ

Agreed. Gross margin is more of an outcome than the source of pricing power. The real question is whether a company can raise prices without volume loss, churn, or substitution. Been thinking about this a lot while building my research tool margins alone can hide very different business realities.

I often meet analysts who claim that the best quantitative measure of pricing power is the gross margin. And while it's true that a relatively high gross margin, more notably the margin stability and gross profitability (gross margin scaled by the assets), can act as an indicator since that would mean the company can charge a lot for its offerings, what the Value Stick framework tells us is that it's not necessarily the case. Two companies currently earning the same gross margin in the same industry can possess vastly different pricing power.

When you finally have full confidence that one day you will be richer than Elon Musk

project link : deep-asset-research.onrender.com

Massively upgraded Deep Asset Research. Now it’s not just a filing summariser it’s becoming a real stock research workspace. New features: • Save reports • Add stocks to watchlist • Get recent news on any stock • Price charts • Revenue + earnings charts • Max drawdown analysis Frankly, this is turning into one of the best tools for stock research. If you have any feedback feel free to reply in this post or DM

Massively upgraded Deep Asset Research. Now it’s not just a filing summariser it’s becoming a real stock research workspace. New features: • Save reports • Add stocks to watchlist • Get recent news on any stock • Price charts • Revenue + earnings charts • Max drawdown analysis Frankly, this is turning into one of the best tools for stock research. Try it: deep-asset-research.onrender.com

Love him or hate him but he is funny af

@DeepPhilo_HQ making a Franklin chatbot for more philosophical ideas like this

"Love your Enemies, for they tell you your Faults". - Benjiman franklin

@michaeljburry *portfolio

@michaeljburry Instead just put your whole phortpolio on $SPCX and let it ride for the rest of your life so that your children might have a ticket to mars

A stock falls from 100 to 10. That's a 90% decline. The investor buys in at 10 because it is indeed worth 30. But it has fallen so much it has become detached from the underlying business and so falls to 5.. The value investor is down 50%. Which isn't as bad as 90%, but is still 50 effing percent. The value investor could have sold at 8, and limited the loss. 10 years later the stock is at $30. While said nvestor gies to medical school and finsihes residency to become a surgeon. That's a six times return in 10 years. About 20% per year. Wait, is it not only a triple? No, because the investor could have sold at 5 like just about everyone else did, but did not. What I describe here is almost exactly WBD. And it didn't take 10 years, because it usually doesn't.

Making a Benjamin Franklin Chatbot for my new project Current learning about RAG and vector embedding. DM/Reply if you wanna collaborate