FΔR

[Roan @RohOnChain]

My friend applied to 200 tech jobs in two years. No MIT. No Stanford.

Last month Anthropic offered him $750,000.

I asked him how he broke in from zero.

He sent me the exact video that got him in. Anthropic's 2 hour course on "How to become an AI engineer in 2026"

Their core team shows exactly how to architect & build AI agents from scratch.

I watched it last night.

Halfway through, I realized I could break into an AI lab in weeks, not years.

Bookmark this and read the article below.

• 00:00 - AI agents with graph engineering
• 06:41 - AI agent architecture
• 14:31 - building AI agent loops live
• 1:15:24 - AI agentic RAG
• 2:20:42 - Anthropic interview process

[Reads with Ravi @readswithravi]

“One day you will wake up and there won't be any more time to do the things you've always wanted. Do it now.”

— Paulo Coelho

[Reads with Ravi @readswithravi]

This sentence from Carl Jung hits hard.

“No matter how isolated you are and how lonely you feel, if you do your work truly and conscientiously, unknown friends will come and seek you.”

[Justin Skycak @justinskycak]

You cannot be creative at a high level unless you are robotic at a low level.

[Sahil Bloom @SahilBloom]

Underrated life advice: A good life is built by simply doing uncomfortable things sooner. The delayed conversation gets harder. The ignored problem gets bigger. The small repair gets more expensive. Peace is never found through avoidance. It’s found through direct early action.

[Afshin Rattansi @afshinrattansi]

With the carnage of the war on Iran, the world has forgotten what US-UK-EU-backed Israel did to Gaza.

Survivors of the genocide live in a hellscape of rubble and ash, which looks like the aftermath of a nuclear war…

And the world looks away.

[Nagma parween @TechByNagma]

INSTEAD OF WATCHING AN HOUR OF NETFLIX TONIGHT.

This 60-minute Cambridge lecture by Demis Hassabis will teach you more about the future of AI than most people will learn in the next 5 years.

Bookmark it and give it an hour, no matter what.

[Sahil Bloom @SahilBloom]

WARNING: Longer post (but worth reading or bookmarking for later).

There's no such thing as too late.

A few days ago, I came across an old newspaper advice column from 2000:

Dear Abby:

I am a 36-year-old college dropout whose lifelong ambition was to be a physician. I have a very good job selling pharmaceutical supplies, but my heart is still in the practice of medicine. I do volunteer work at the local hospital on my time off, and people tell me I would have made a wonderful doctor.

If I go back to college and get my degree, then go to medical school, do my internship and finally get into the actual practice of medicine, it will take me seven years! But, Abby, in seven years I will be 43 years old. What do you think?

- Unfulfilled in Philly

Dear Unfulfilled:

And how old will you be in seven years if you don't go to medical school?

Let that one sink in for a moment.

That timeline in your head is your own creation.

It forms from a young age. We look out at the world and use what we observe to sketch out a map of how our life is supposed to go. Social media highlight reels tell us how we stack up. What we should be doing. By when. And what we can’t possibly do now.

It’s a form of inheritance. Not one we choose, but one we all get.

So, we tie ourselves in knots, just like Unfulfilled in Philly. We know what we really want, but we’re too afraid to go all in. Because to go all in is to expose yourself to real failure. To be completely naked.

But the harsh truth that Abby so elegantly highlighted is an important one:

Time is unforgiving. It doesn’t care about you. It’s moving with or without you. Whether you choose to act…or not. Whether you live the way you want…or not. Time is just going to pass.

How often do we all sit around waiting for someone else to come tap us into our own life? To give us permission to go and do the thing we want to do?

Waiting for our own Dear Abby reply to tell us it’s ok. We’re not too late. We’re right on time.

The real lesson here is recognizing that you don’t need any of that. You don’t need this post. You don’t need Dear Abby.

You get to tap yourself in. You get to choose.

You're not late. You're not early. You're exactly where your life has led you to be. And you get to decide where you go from here.

So, that thing you want to do:

How old will you be in ten years if you don’t do it?

[Burcun Ne? @NeBurcusun]

Fizik tedavi hocam bana bunu söyledi ve bunun öne eğik boynu (text neck / turtle neck) düzeltmek için en iyi yöntem olduğunu söyledi.

Bir hafta boyunca uyguladım ve boynumdaki tüm sertlik tamamen kayboldu. Hatta o kadar alışmışım ki, sertlik gidince garip hissettim.

Her gün 10 dakika boyunca bu şekilde yüzüstü uzanmayı deneyin.

Bunu yaptığınızda ensenizdeki o gerginlik ve sertlik hissi kayboluyor. 😱

[Brian Maierhofer @brianmaierhofer]

The most attractive human trait is energy. Full stop.

You'd think it was looks or status, but your nervous system reads deeper. It's called neuroception, the body's pre-conscious scan for enthusiasm, curiosity, and safety. Want to become a magnet for exceptional people? Focus on your energy. Embody the kind you want to attract.

[Aporia @aporia9n]

In fact half the people I meet in beach clubs in St Tropez / Marbella are doing exactly this: one year it’s crypto, then an Amazon brand, then an AI lead-gen agency, then solar panels or some subsidised renovation business.

Make €1m here, €2m there, shut it down when the margins disappear, spend six months travelling, then come back with the next “coup.”

I used to think this was just unserious entrepreneurship, but some of them have become genuinely very good at it. Their real business isn’t crypto, AI or solar panels. It’s being early.

They spot a new distribution channel, regulatory distortion, shortage or wave of hype before everyone else, build something around it very quickly, then leave before the market gets crowded.

There’s no attachment to the product or industry because the company is only the vehicle. What actually compounds is the cash, contacts, execution speed and instinct for where the next trade will be.

The old founder wanted to build one company that would outlive him.

The new one wants enough wins that he never has to depend on one company at all.

[ثيو @egybest_1]

مايك تايسون، في عز مستواه، كان يسوي إلى ٥٠٠ تكرار من هذي الحركة، المعروفة باسم "تمارين تايسون".

هي من أقوى تمارين وزن الجسم، لأنها تشغل الجسم كامل.

وتقدر تمرّن الرجلين والجزء العلوي من الجسم في نفس التمرين.

وإليك روتينه في التمرين ..

[Richard Medhurst @richimedhurst]

Knowing how to defend against state surveillance should be mandatory for journalists in 2026.

Even with it, you're already at a disadvantage against state actors. Without it, you risk lives and sources.

That's exactly why I'm writing a massive cybersecurity bible.

[Reads with Ravi @readswithravi]

The most powerful passage from Meditations by Marcus Aurelius:

[Machiavelli Bot @UnmodernmanBot]

The average man is too honest in places where he should be strategic, and too dishonest in places where he should be accountable. He reveals his plans to strangers, but lies to himself about his habits. Power begins when you reverse this. Be private with the world. Be mercilessly honest with yourself.

[Paul Graham @paulg]

What's interesting about these questions is the ratio of how important they are to how often you think about them.

Every company probably could and should do its own version of this. In fact it would be a useful exercise just to decide what the questions are.

[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.

[Vaibhav Sisinty @VaibhavSisinty]

A guy on Reddit with 10 years of engineering experience just shared the one thing he'd teach every vibe coder first.

And it'll save you thousands in AI costs. 🤯

Most people using Claude Code use it the expensive way. They call the AI every time the tool runs. Every run burns tokens. Every token costs money.

His advice: flip it. Use Claude Code to BUILD the tool once. Then run it forever without spending a single token.

Simple example. You want to check a website daily for updates. The expensive way: have an LLM search the site every day. Burns tokens every single time.

The free way: use Claude Code to write a script that scrapes the page and alerts you if anything changed. Build it once. Runs forever. Zero tokens.

Then he took it further. He had Claude Code build him a full neural network something that used to take weeks and years of ML training while he cooked dinner.

It runs for free. No tokens. No API calls. Forever.

Spend tokens once to build it. Run it for free forever.
That's it. That's the insight most vibe coders are missing.

[ℏεsam @Hesamation]

this Anthropic researcher wrote one of the best articles you can find here on AI research. obviously, Hamming's classic book is highly suggested to learn how to develop a "Research Taste".

[Amjad Masad @amasad]

This is the most inspiring positive-sum vision for AI in the enterprise.