Darshil Dixit

Seeking Truth

[Anatoli Kopadze @AnatoliKopadze]

Karpathy just said the people who don't use LLMs are already losing

he spent 4 minutes explaining why smart people are still going to fall behind

not only the people who refuse AI, but also those who think signing up for Claude counts as using it

typing a prompt and reading what comes back isn't the skill, anyone can do that

the real shift is going from asking AI things to building something that runs without you

that's exactly why I wrote an article on Claude features most people don't even know exist

read the article below and you'll already be ahead of 99% of people

[GREG ISENBERG @gregisenberg]

The most valuable skill sets on the planet right now:

1. people who can set up agents properly, manage them, and run local AI models
2. marketers who know how to build distribution
3. robotics engineers who can do all three: build the hardware, wire in the AI, and source manufacturing etc
4. curators who are good at yapping and can do short form video in their sleep
5. the builder-distributor. The one person who can both ship the product AND get it in front of people
6. IRL community builders

[darkzodchi @zodchiii]

NVIDIA CEO, Jensen Huang:

"Nobody writes prompts anymore. The new job is to write and handle loops."

He calls it the shift that defines the rest of 2026.

Interview was out just yesterday.

Watch the 23 minute talk, then save the full framework below👇

[Vaibhav Sisinty @VaibhavSisinty]

Vercel cooked something genuinely special here. 🤯

They open-sourced the exact framework they use to run 100+ AI agents internally. And the way it works changes how you think about building agents.

It's called Eve. An agent is a folder. Tools are files. Skills are markdown files. Channels are files. The folder structure IS your agent.

One command to start:
npx eve@latest init my-agent

No plumbing. No boilerplate. Eve handles durable execution, sandboxed compute, human approvals, evals, tracing, and deployment all built in.

Add a tool? Drop a TypeScript file. Add a skill? Drop a markdown file. Add Slack? One command. Add a schedule? One more file. Deploy it? vercel deploy.

How Vercel already runs on Eve:

→ Data analyst agent handles 30K+ questions per month in Slack

→ Sales agent costs $5K/year and returns 32x that

→ Support agent solves 92% of tickets on its own

→ 29% of all Vercel deployments now come from agents

Their bet: Next.js ended the era of hand-rolling websites. Eve ends the era of hand-rolling agents.

[darkzodchi @zodchiii]

Google CEO, Sundar Pichai:

"If you don't learn to how to orchestrate agents now, you'll spend 2027 catching up to people who started today"

In 30 minutes he explains why the best engineers stopped writing code and started running agents.

Watch the interview, then save the exact setup below 👇

[Alex Vacca @itsalexvacca]

https://t.co/31dGiQf8Np

[Dan Rosenthal @dan__rosenthal]

https://t.co/VwnHkqCluJ

[Rahul @sairahul1]

This is the best site on the internet to learn how LLMs actually work.

Free. Completely.

https://t.co/YOGF6PsmBN

Bookmark this site.

Then read this setup ↓

[Rahul @sairahul1]

Anthropic pays $750,000+ a year for engineers who can build LLMs from scratch.

Not how to prompt them.
Not how to fine-tune them.
Not how to build RAG pipelines.

But how to build them from scratch.

This 2-hour Stanford lecture teaches you everything.

Scaling laws.
Data collection.
Architecture design.
Post-training alignment.

Free. From Stanford.

Watch first. Then read this.

The lecture is the theory.

And this article shows you how to actually build it (with code) ↓

[Varun Mayya @waitin4agi_]

I really like this article. I think that the capabilities of a country are fully dependent on the local buying capacity and the size of the local economy. Whether physical or digital, the farther away you have to distribute something, the more expensive it is. Local distribution is always much easier but if local buying power is low then companies have to export which carries a bigger distribution cost because you have to compete against locals in other countries who have a much lower distribution cost. Actually it’s not just distribution: it’s everything from consumer insights to feedback to key relationships that are a distance away. So if you do innovate but you have to necessarily export that innovation to make money, you’re at a disadvantage against local players.

One personal example is that India has never made high quality games simply because the local purchasing power is low. If we had a lot more PCs things would be very different. China has roughly 320 million PC gamers and India is about 39 million. So on players alone, China’s PC base is roughly 8x India’s. BUT China’s PC game spend is on the order of 80–100x India’s, even though its player base is only like 8x larger. The difference mainly is monetization as Indian gamers spend much less (core ARPU has run around $0.29/month), so 39M PC players translate into very little premium game revenue. Game Science’s art director Yang Qi confirmed that nearly 70% of Wukong’s sales came from China itself. Knowing a local buying market exists justifies spending. The only way we can justify what we are spending now on UTA is because we found inroads into global markets through content otherwise this would be a money losing exercise.

The other problem is that low purchasing power economies have too tiny a market for early adopters. If you built an OpenAI in India before anyone else 50% of people wouldn’t believe you and 50% of people will tell you it won’t work or doesn’t have use cases. I think you need a crackpot high purchasing power early adopter network with high failure and bullshit tolerance to make truly innovative things and also forgive crazy companies during early mistakes because history teaches us that the best companies all had v0.1s that were not very convincing to the masses.

Thats why it’s critical for anyone who wants this country to succeed to first really create more jobs, more disposable income, even if that means creating the nth packaged food brand (American grocery stores still have a much wider variety of biscuit brands than India for example) or food delivery apps before they take bigger bets. Not because they need the capital themselves to try bigger bets, but so that they can diffuse more capital into the ecosystem via jobs and the rewards of equity ownership such that that cohort of people become early adopters for other innovative companies.

Success comes from satisfying local market demand (sometimes like in the case of Tesla or Ford there is hidden demand and entrepreneurs need to unlock it) and rarely comes from creating something that has no local demand. After studying Chinese social media so much I have a long thesis on why they did well (bans on global social media platforms constrain desire of products to local players only who now get revenue and profit to do RnD. Think about what % of disposable income from India is being spent on global brands where the desire to buy starts on a global social media platform).

Anyway people complaining about India building “easier businesses” are really not spending the mental energy to think second order. And 9/10 times this same type of person will completely ignore local innovation that is almost always happening in parallel but gets less media coverage.

[GREG ISENBERG @gregisenberg]

Fable is banned. Long live local AI.

Full episode breaking down exactly how to get good at local models. the runtime, the hardware, quantization, connecting it to Hermes agent and local AI startup ideas (25 minutes)

[GREG ISENBERG @gregisenberg]

The takeaway from Fable 5 being BANNED by the government: GET GOOD AT LOCAL MODELS SO YOU HAVE 100% CONTROL.

My entire weekend was going to be building my craziest ideas with Fable 5. That's now cancelled.

So instead of building with Fable this weekend, I've decided I'll go deep on local models:

1. Start with the runtime. Download Ollama or LM Studio first. This is the thing that actually runs models on your machine.

2. Match the model to your hardware. A model's size is measured in billions of parameters (7B, 32B, 70B). Bigger is smarter but needs more memory. Rule of thumb: a 7B model runs on almost any laptop, a 32B needs a good Mac with 32GB+ RAM, a 70B needs serious hardware like a DGX Spark or a maxed-out Mac Studio.

3. Know which model for which job. Qwen 3 is the best all-around choice for most tasks. DeepSeek for reasoning and coding. Gemma 4 when you need something tiny that runs on a phone. Llama when you want the biggest community and the most fine-tunes.

4. Quantization. You can shrink a model to run on weaker hardware with barely any quality loss. Look for versions labeled Q4 or Q5. This is how a model that "needs" a server runs on your laptop. Learning this one concept changes everything.

5. Connect it to your agent. Point Hermes or your agent stack at a local model.

6. Context window is your real constraint locally. Cloud models give you huge context for free. Local models make you pay for it in memory. A bigger context window eats RAM fast. Keep your sessions tight and your prompts lean or your machine chokes.

7. Learn to give local models tools. A smaller local model with web search, file access, and code execution beats a giant model with none. The capability gap closes fast when you wire up the right tools. The model is the engine but the tools are the wheels.

8. Fine-tuning is more accessible than you think. You don't need this on day one, but know it exists. You can take an open model and train it on your own data so it gets good at your specific domain.

I'll probably do a breakdown at some point on this @startupideaspod if people are into it.

The lesson from this ban is basically don't build your entire workflow on something that can disappear with a single letter. Own part of your stack. Local models are insurance.

It reminds me when people realized they don't own social media accounts. And then you saw people build email lists etc.

I remember running a startup and my biggest traffic source was organic FB. All of a sudden, algo changed, and I lost 99% of my traffic.

Same sorta moment (but bigger) for AI.

This is a wake up call.

[Darshil Dixit @Darshil_Dixit]

@elonmusk @yatharthmann @LorenaSGonzalez @grok what was the context of this whole tweet?

[Alex Finn @AlexFinn]

AI subscriptions are dead

Claude Fable 5 will only be on the Anthropic subscription until June 22nd. After that, you will need to pay for usage per token

This will be the start of a much larger trend

Frontier models will no longer be included in subs

You’ll pay a fee and it will only get you access to older, much cheaper models

If you want access to that dank AI sour diesel, you’re going to need to pay for every token you use. No more subsidies

And it make sense. The subsidies were just a Ponzi scheme

For those that don’t know, when you pay $200 a month for an AI sub, you get thousands of dollars of tokens

These AI companies actively lose tremendous amounts of money because of these subscriptions. GDPs of most countries every year are lost on your $200 Claude Max sub

The investor money is running dry. IPOs are coming because of this. And with IPOs need to come profitability

The golden age of paying $200 a month and being able to code on 40 Claude Code instances and getting a usage reset every 5 minutes are about to die

The party couldn’t continue ever. You can’t just leverage the entire global economy for years and expect nothing to break. Now it’s time to pay up

Means a few things:

1. Time to be responsible when it comes to which models you use. You don’t need Fable 5 for GPT 5.5 Xhigh for everything. Build the skill of knowing when to use cheap models

2. Local LLMS/hardware will come even more in demand. I’m currently running GLM on my Mac Studio. It’s great. Is it Fable? No. But it gets the job done for free on simple tasks. Learn about local LLMs

3. This is the beginning of the wealth gap expansion. Those that can afford to spend $10,000 a month on Fable 5 will build incredible products that eat up more and more of the economy. Those that can’t afford Fable 5 will have an insane disadvantage

4. The government will need to step in eventually. There will be too much civil unrest. I hope the answer isn’t free money. That won’t do anything. I hope the answer is education/access to AI resources for ALL. Universal Basic Opportunity

5. You need to seriously reconsider where your money goes every month. If you are complaining about AI prices and in the back of your mind you know your skill set is becoming quickly irrelevant, all while spending money every month on Netflix, Xbox Live, Paramount +, drugs, DoorDash, Uber, and other things that bring nothing positive to your life, you are simply doing it wrong. AI is an investment in yourself. It’s an investment in your relevance to the global economy. You need to make sure you make that investment

The pieces on the board are quickly moving around. The rules are changing. The battlefield is shifting. If you’re not strategizing accordingly, you’re cooked.

[Vaibhav Sisinty @VaibhavSisinty]

A Director at Google just explained the difference between using AI like a calculator and using AI like an employee.

The employee version has a name. It's called Loop Engineering. And once you understand it, you'll never use AI the same way.

Loop Engineering means you stop steering and start building the road.

You design a system that gives AI work, checks its output, remembers what's done, and moves to the next task. Automatically.

Think of it like this. Cooking dinner every night is prompting. Building a kitchen that cooks on a schedule is loop engineering.

Here's what's actually inside a loop:

→ Automations : a timer that wakes AI up to find work on its own

→ Worktrees : separate desks so multiple agents don't step on each other

→ Skills : your project knowledge written down so AI stops guessing

→ Connectors : wires to your real tools: Slack, GitHub, databases

→ Sub-agents : one AI writes, a different AI reviews

→ Memory : a file that tracks progress because AI forgets everything between runs

This framing comes from Addy Osmani spent 14 years leading Chrome DevTools, now a Director at Google Cloud AI working on Gemini.

But his biggest point isn't about the loop. It's about you.

Two people can build the exact same loop. One uses it to go faster on work they deeply understand. The other uses it to avoid thinking entirely.

The loop can't tell the difference. You can.
Build the loop. Stay the engineer.

[Vaibhav Sisinty @VaibhavSisinty]

Everything you need to master Claude Code is free, and sitting on the internet right now. 😨

Most people will never find these. So I pulled together the 5 that actually matter:

𝟏. 𝐂𝐋𝐀𝐔𝐃𝐄.𝐦𝐝
The one file that gives Claude memory. Drop it in your project root and it reads it at the start of every session. No more re-explaining your stack, your decisions, or your preferences every single time.
🔗 https://t.co/mEWPJgAWen

𝟐. 𝐒𝐤𝐢𝐥𝐥𝐬
Teach Claude a reusable workflow once, and it repeats it forever. You build it one time, Claude runs it on demand.
🔗 https://t.co/pPmPoDBEHh

𝟑. 𝐌𝐂𝐏
Connect Claude to Slack, GitHub, and Google Drive in one setup. This is the piece that turns Claude from a chatbot into a tool that actually works inside your stack.
🔗 https://t.co/CSmAlqqPcT

𝟒. 𝐑𝐨𝐮𝐭𝐢𝐧𝐞𝐬
Automate tasks to run 24/7 without you touching anything. Hand Claude a recurring job, it runs on schedule, and you wake up to finished work.
🔗 https://t.co/Bmv0QM7A64

𝟓. 𝐂𝐥𝐚𝐮𝐝𝐞 𝐂𝐨𝐝𝐞 𝐔𝐥𝐭𝐢𝐦𝐚𝐭𝐞 𝐆𝐮𝐢𝐝𝐞
The best community-built resource for everything else, hooks, plugins, advanced workflows, all in one repo.
🔗 https://t.co/9mlV47YY7j

All 5 are free. All 5 are live right now.

The people who learn this stack today will look back in 18 months at everyone who didn't the way founders look at people who refused to use the internet in 2001.

Save this post. You'll come back to it.

[GREG ISENBERG @gregisenberg]

What does it actually mean to be AI native?

There was no clear guide on the internet for how to become AI native so we built the definitive one (60 min masterclass):

1. An AI native org has 3 layers: people for strategy and taste, agents for execution, and a shared context layer that makes the entire company readable to agents.

2. AI eats the middle of your work. You used to spend 80% of your day on execution. Now agents do that. Your job is the bookends: deciding what to do and judging whether it's good enough.

3. Everyone is a manager now. Your output is the output of your agents. If your agents produce garbage, that's on you. You set them up wrong.

4. Using ChatGPT doesn't make you AI native. That's like having a website and calling yourself a tech company lol.

5. No AI native org without AI native people. Most companies skip straight to the tools. That's why it fails. If your people don't understand how to manage agents, the tech doesn't matter.

6. Making your company "readable" to agents is the real work. Every process, every decision, every piece of knowledge needs to exist in a format an agent can consume. Most companies are nowhere close.

7. Speed without signal is just expensive chaos. You need the system to move fast AND know if you're moving in the right direction.

8. The skill chain is how agents get good at your specific workflows. Skills build on skills. The more you invest in them, the more your company compounds.

9. The moat is the system. People managing agents, agents reading from rich context, the whole thing getting smarter every week. That compounds. Your competitor can copy your tools. They can't copy your system.

Full episode with @TheoTabah from @meetLCA on @startupideaspod. This is the stuff we normally keep internal but all the sauce is yours.

@TheoTabah is the brains behind advising the world's biggest companies on AI and building AI products. Your fav CEO's first call for figuring out AI.

You are in for a treat

Become AI native in under 60 minutes

https://t.co/EzreBHFyIJ

Watch

[Vaibhav Sisinty @VaibhavSisinty]

Okay this is genuinely insane.

SpaceX just unveiled a satellite whose only job is to run AI. Not internet. Not GPS. Just compute, floating in orbit.

It's called AI1, and the reason behind it breaks your brain.

AI data centers on Earth are hitting a wall, not a chip wall, a physics wall.

They need staggering amounts of power and water just to stay cool, and we're running out of grid and land to build them.

So Musk's answer is: stop building them on Earth.

In orbit, the sun never sets. Free power, 24/7. No water for cooling, you just radiate heat into the vacuum of space. The two things choking AI on the ground barely exist up there.

And here's the wild part: Musk says it's easier to build than a Starlink satellite. Strip out the complex antennas and it's "a lot of solar cells, a radiator, and some laser links."

One AI1 carries the compute of an Nvidia GB300 rack, the same hardware data centers fight over down here.

AI1 is just the first one. The plan is a constellation of up to a million of them.

And the timing isn't an accident, SpaceX goes public this week at a ~$1.75 trillion target. This isn't a rocket company anymore. It's positioning itself as the power grid for AI, in space.

The race for AI compute just left the planet. Literally.

@SpaceX

[Rahul @sairahul1]

Boris Cherny, Head of Claude Code at Anthropic:

"The next transition is coming this year."

2023: you wrote the code.
2024: you prompted Claude to write the code.
2025: you wrote loops that prompted Claude.
2026: you build the harness that runs the loops.

Each transition, the engineers who understood the new abstraction first moved ahead permanently.

The ones who waited caught up eventually — but never all the way.

Harness engineering is the 2026 abstraction.

Here is everything you need to know ↓

[Rahul @sairahul1]

Jensen Huang, CEO of Nvidia:

"Every engineer is going to have and manage hundreds of agents."

The most valuable engineering skill of 2026 is not taught in any university.

No CS program teaches harness engineering.
No bootcamp teaches agent memory architecture.
No degree prepares you to build systems that survive production.

One builder mapped the entire thing out — free, step by step, no degree required.

This is the roadmap ↓

Bookmark this for the weekend.