Dirk Liebich

Entrepreneur, Executive, Digital Architect, Author, Blogger, Public Speaker, and Founder iCUE

[Dirk Liebich @Dirk_Liebich]

Why is Claude code so dumb all of the sudden? None of the models really do its job, all are skipping steps, are slow, and lying. It’s so bad, I’m considering going back to Codex. This is redic.

[Dirk Liebich @Dirk_Liebich]

@vipul_1011 Interested

[Dirk Liebich @Dirk_Liebich]

@honeymoon250 Is that even a question?

[Dirk Liebich @Dirk_Liebich]

@Brink_Thinker Love this. God bless you

[Eyal Toledano @EyalToledano]

https://t.co/QtzAUip9Vu

[Dirk Liebich @Dirk_Liebich]

@Camp4 Strongly agree! Hbd

[Kevin Dahlstrom @Camp4]

Today I turn 55.

I’m the fittest, sharpest, and happiest I’ve ever been.

If I’m an outlier, it’s not because I’m built different or discovered a secret formula. The truth is far less glamorous:

It’s a million tiny choices, compounded over decades.

Here are 55 of them:

1. Walk 15+ miles a week, even if you do other exercise. Humans are uniquely made to move slowly over long distances—it’s critical to longevity.

2. Develop a writing practice. It’s the single best way to sharpen your mind. And remember, you don’t have to be a good writer to write. Start with 10 minutes a day.

3. Swap out your toothpaste, deodorant, lotions, soap, shampoo, and other personal care products for natural versions. Here’s a rule of thumb: Don’t put anything on your skin that you couldn’t safely eat.

4. If you have a positive thought about someone, don’t keep it to yourself—share it immediately. Encouragement defies the laws of physics: When you give energy, you also receive it.

5. Wear shoes with a wide forefoot (I like Topo Athletic) and wear toe spreaders around the house (search “yoga toes” on Amazon). Spine health begins with the feet.

6. Get sunlight regularly. Moderate sun exposure (without sunscreen) is hugely important for overall health.

7. Do a 3-minute deep (“ass to grass”) squat every morning. Deep squats are often called the anti-aging exercise. It’s been said that, “It’s not that you can’t do deep squats because you’re old, it’s that you’re old because you can’t do deep squats.”

8. Explore minimalism (it’s not what you think it is).

9. Set boundaries on toxic relationships. We tend to cling to relationships past their expiration date, and it takes a bigger toll on our health than we recognize.

10. Eat real food. Not too much. Don’t eat garbage. Binge occasionally. Fast occasionally. That’s the diet.

11. Learn about FIRE. It’s a great framework for financial success.

12. Don’t take antibiotics except in emergency situations. They’re massively over-prescribed and aren’t needed in most cases. Antibiotics have done untold damage to our guts, which is where health begins. Great natural alternatives are out there.

13. Get 8 hours of quality sleep each night. To optimize sleep:
—Don’t eat after 6pm
—Get blackout shades and cover LEDs with black tape
—No screens 2 hours before bed
—Try ashwagandha (an herb) to calm the nervous system

14. Stop drinking, even in moderation. People find all sorts of ways to justify drinking, but there’s no escaping the simple fact that alcohol is a toxin and it limits your potential.

15. Travel as much as possible. Nothing expands the mind like seeing the world. And travel doesn’t have to be expensive—the best experiences happen outside of fancy resorts, when you live like a local.

16. Let go of resentment. When you forgive someone, you release the prisoner, and the prisoner isn’t them… it’s you.

17. Show up on time, every time. Poor time management limits success more than most people realize. If you struggle with punctuality, stop everything else and fix that first.

18. Spend lots of time in nature and touch the earth. Humans evolved over 300k years to live in harmony with nature, and only recently have we retreated indoors. If you don’t spend time outside, you’re fighting biology (hint: You won’t win.)

19. Stop doing dumb things. As Leo Tolstoy said, “People try to do all sorts of clever and difficult things to improve life instead of doing the simplest, easiest thing—refusing to participate in activities that make life bad.”

20. Find your happy place and (eventually) move there. Most people live where they live because... that's where they live. We are products of our environment—choose yours carefully.

21. Find a hobby and pursue mastery. You can’t have a happy life without a passionate pursuit that isn’t your vocation. Your work—even if you enjoy it—isn’t enough.

22. Avoid mainstream medicine except as a last resort. The results are in—our healthcare (or more appropriately, sick care) system is badly broken and only makes people sicker.

23. Have a mindset of abundance. There is no advantage to being a pessimist—even if you’re right, it’s a miserable way to live. In a very real way… whatever you believe, you’re right!

24. Do hard things. Choose courage over comfort. Everything you want is on the other side of fear and hard work. As Jerzy Gregorik said, “Hard choices, easy life. Easy choices, hard life.”

25. Ignore haters. Hurt people hurt people. Negative/toxic people live in a prison of their own design. Don’t join them!

26. Say no. Protect your time and energy like it’s your most precious asset… because it is.

27. Become a water snob. As an alien said on Star Trek, humans are “ugly bags of mostly water.” You are what you drink—literally! We have Mountain Valley Spring water delivered in glass 5-gallon jugs and also have whole-house water filter (Aquasana Rhino).

28. Stop drinking sodas and sugary energy drinks. After a few weeks you won’t miss them, and a few months later they’ll seem disgusting. Refined sugar causes inflammation, which is the root of most disease.

29. If you’re over 35, find a good functional/longevity medicine doctor and start tracking your hormones. Modern life is hell on the endocrine system and restoring healthy hormone levels can change your life. As we get older, we either accept a slow decline in performance or we do something about it—choose the latter!

30. Develop a morning routine and follow it faithfully. Win the morning, win the day!

31. Invest in experiences, not things. People frequently regret buying things, but rarely regret investing in great experiences (especially when shared with loved ones). Remember, there’s nothing you can buy in a mall that you’ll remember in ten years.

32. Explore spirituality. It’s arrogant and small-minded to believe there’s nothing going on in our universe that is beyond our comprehension. We know less about our universe than an ant meandering on a sidewalk understands about this planet.

33. Have a strong bias toward action—doing rather than talking. If you ask a bunch of old people about their regrets, they’ll talk about the things they *didn't* do—the shots they didn’t take—more than the things they did do (even if it went wrong). As Wayne Gretzky famously said, “You miss 100% of the shots you don’t take.” Most people don’t take enough shots.

34. Stay lean. Men in particular are obsessed with muscle mass these days, but bulk doesn’t age well. The goal is to be strong but lean. The fittest guys in their 50s and beyond aren’t meatheads, they’re lean guys who are serious about a sport.

35. Curate your inner circle carefully. Surround yourself with people you admire and who challenge you to grow. Remember, we’re the average of our 5 closest relationships.

36. Be the fittest version of yourself. Your body is your only vessel for experiencing life—so treat it as such. Fitness isn’t working out a few times a week, it’s a lifestyle. The older you get, the more time you need to devote to your health.

37. Take the time to appreciate art and beauty in all its forms.

38. Think globally, but act locally. Too many people put their energy into far-away problems they don’t understand and can’t impact, while ignoring problems right under their nose. Want to change the world? Start at home.

39. Try psychedelics. It’s one of those things everyone should do at least once, and it might be the breakthrough you’ve been looking for.

40. Limit bad habits, including unhealthy thought patterns. We all have them—practice avoidance and find substitutes. Get professional help if needed.

41. Be a lifelong learner. Your brain is just like a muscle—if you don’t feed and flex it regularly, it will atrophy.

42. Find your purpose. People with a strong sense of purpose are happier and live longer. Lack of purpose sucks energy and magnifies depression.

43. Only take advice from people who embody the traits you want to have. Talk is cheap—emulate those who have DONE it.

44. The goal is not to retire and do nothing, it’s to build a great day-to-day life that you don’t need to escape. A life of leisure is a slow death. Happiness isn’t possible without a little struggle, uncertainty, and skin in the game.

45. Have fun! Do frivolous and silly things that make you smile. As George Bernard Shaw famously said, “We don't stop playing because we grow old; we grow old because we stop playing.”

46. Whatever you want to do or achieve in life, start NOW. Don’t fall victim to “someday thinking” because someday never comes.

47. Accumulate assets—things that grow in value over time. It’s the #1 habit of rich people, and it can be done in tiny chunks. Instead of spending $100 on an impulse purchase that has no lasting value, put that money into an index fund or Bitcoin. It becomes addictive (in a good way).

48. Don’t ignore the big 3 canaries in the coal mine for health:
—Low libido (and ED)
—Frequent sinus & respiratory issues
—Depression
These usually aren’t medical conditions in themselves, they’re symptoms of an underlying problem. Find a good doc (outside of the mainstream) and figure out the root cause.

49. Have a clear vision for your future. How can you decide which direction to go if you haven’t clearly defined the destination? It sounds obvious, but 95% of people haven’t defined their “Ideal End State” in detail and in writing. (Check out my thread on this topic.)

50. Make your own decisions. We live in an era where most of what society tells us is wrong. Don’t be afraid to break from societal norms—if people say you’re crazy, it’s a sign that you’re doing something right.

51. Get hardcore about mobility exercise. As you age, it’s usually the knees, hips, and lower back that limit physical performance. 30 min a couple times a week can spare you a lifetime of pain. YouTube is a great resource.

52. Go all in on family. Get married, stay married, have kids. Burn the boats. In the end, family is all that matters.

53. Be ruthless with your time. Money comes and goes. Time only goes. Audit your calendar ruthlessly—cut the trivial, double down on the meaningful, and spend your hours like your life depends on it. (Because it does.)

54. Have a strong bias toward action. Be curious, try things, meet people—it’s how you increase your surface area for serendipity, the most powerful unseen force in our lives.

55. Reinvent yourself every decade. Over time, we slowly drift off course from our priorities, values, and true identity. Take stock and don’t be afraid to hit the reset button. Bold, calculated moves made for the right reasons almost always pay off—usually even more than you can imagine.

🎁 P.S. If you enjoyed this post, would you give me a birthday gift? Repost or comment with the item number(s) you liked best?

[Luyen Chou @luyenchou]

@karpathy As an educator and long-time edtech champion I completely agree. But I also think new AI tools can greatly help educators understand and improve student learning at scale - far better than standardized tests and traditional assessment! https://t.co/0AbKV5GCkq

[Dwarkesh Patel @dwarkesh_sp]

The most interesting part for me is where @karpathy describes why LLMs aren't able to learn like humans.

As you would expect, he comes up with a wonderfully evocative phrase to describe RL: “sucking supervision bits through a straw.”

A single end reward gets broadcast across every token in a successful trajectory, upweighting even wrong or irrelevant turns that lead to the right answer.

> “Humans don't use reinforcement learning, as I've said before. I think they do something different. Reinforcement learning is a lot worse than the average person thinks. Reinforcement learning is terrible. It just so happens that everything that we had before is much worse.”

So what do humans do instead?

> “The book I’m reading is a set of prompts for me to do synthetic data generation. It's by manipulating that information that you actually gain that knowledge. We have no equivalent of that with LLMs; they don't really do that.”

> “I'd love to see during pretraining some kind of a stage where the model thinks through the material and tries to reconcile it with what it already knows. There's no equivalent of any of this. This is all research.”

Why can’t we just add this training to LLMs today?

> “There are very subtle, hard to understand reasons why it's not trivial. If I just give synthetic generation of the model thinking about a book, you look at it and you're like, 'This looks great. Why can't I train on it?' You could try, but the model will actually get much worse if you continue trying.”

> “Say we have a chapter of a book and I ask an LLM to think about it. It will give you something that looks very reasonable. But if I ask it 10 times, you'll notice that all of them are the same.”

> “You're not getting the richness and the diversity and the entropy from these models as you would get from humans. How do you get synthetic data generation to work despite the collapse and while maintaining the entropy? It is a research problem.”

How do humans get around model collapse?

> “These analogies are surprisingly good. Humans collapse during the course of their lives. Children haven't overfit yet. They will say stuff that will shock you. Because they're not yet collapsed. But we [adults] are collapsed. We end up revisiting the same thoughts, we end up saying more and more of the same stuff, the learning rates go down, the collapse continues to get worse, and then everything deteriorates.”

In fact, there’s an interesting paper arguing that dreaming evolved to assist generalization, and resist overfitting to daily learning - look up The Overfitted Brain by @erikphoel.

I asked Karpathy: Isn’t it interesting that humans learn best at a part of their lives (childhood) whose actual details they completely forget, adults still learn really well but have terrible memory about the particulars of the things they read or watch, and LLMs can memorize arbitrary details about text that no human could but are currently pretty bad at generalization?

> “[Fallible human memory] is a feature, not a bug, because it forces you to only learn the generalizable components. LLMs are distracted by all the memory that they have of the pre-trained documents. That's why when I talk about the cognitive core, I actually want to remove the memory. I'd love to have them have less memory so that they have to look things up and they only maintain the algorithms for thought, and the idea of an experiment, and all this cognitive glue for acting.”

[Tom Dörr @tom_doerr]

https://t.co/WnnXIMLIyE

[Orit Peleg @oritpeleg]

More on collective behavior: Our new Annual Review of Biophysics piece - with the stellar Danielle Chase - explores how animals sense, share information, and make group decisions. In honeybees and beyond 🐝

https://t.co/UcuG35gUu5

[Dirk Liebich @Dirk_Liebich]

@Saboo_Shubham_ Not working very well at all

[Nirant @NirantK]

Apple just dropped a killer open-source visualization tool for embeddings — Embedding Atlas — and it’s surprisingly powerful for anyone working with large text+metadata datasets.

This reminds me of Nomic's Atlas, but I never got around to using it 😅

We’re talking real-time search, multi-million point rendering, and automatic clustering with labels.

One of their showcase examples visualizes ~200K wine reviews using embeddings + metadata like price, country, and tasting notes. And it is lightning fast even on my browser! No separate code needed!

It nails what most LLM devs need but often hack together:

✅ UMAP projections
✅ Faceted search across metadata (e.g. “country vs. price”)
✅ Hover + tooltip on raw points
✅ Interactive filters, histograms, and cluster overlays
✅ Cross-linked scatterplot + table views

Under the hood:
• Fast rendering using WebGPU (with WebGL fallback)
• Embedding-based semantic similarity search
• Kernel density contours for spotting clusters or outliers

You just upload your .jsonl or .csv with text + vector + metadata. It handles the rest: clustering, labeling, UI layout, everything.

This feels like the LLM-native version of Tableau — but optimized for text, chat and modern data needs

If you’re building RAG evals, search tuning, clustering explainability, or even dataset audits — this could be your new favorite tool.

[Dirk Liebich @Dirk_Liebich]

@jeremyphoward I've to agree! Same experience here. Sorry, but gtp5 is not really cutting it and slow on top.

[Clint Jarvis @clinjar]

Stanford paid 35,000 people to quit social media.

This was the largest study on emotional health in history.

The results were so shocking, scientists called it "comparable to therapy."

Here's what happens when you break free from the algorithm:
🧵

[Philipp Schmid @_philschmid]

Fuck Yes! Serverless GPU for everyone! The Cloud Run just shipped Serverless GPU with no quota request required. 🤯 Deploy @GoogleDeepMind Gemma with a single command!

- Pay-per-second GPU billing.
- Scale to zero instances.
- TTFT of 19 seconds for a Gemma3 as cold start.
- No quota request needed for L4 GPUs. (3 per default).

[. @jdpoc]

Today, in Tottenham, North London.

[Brian Roemmele @BrianRoemmele]

Magnetic micro robots. They can navigate through your blood vessels.

[Matthew Berman @MatthewBerman]

Anthropic just dropped an incredible guide on "How To Build Effective Agents"

2025 will be the year of AGENTS 🤖

Here's everything you need to know: 🧵

[Dan McAteer @daniel_mac8]

🔥 💻 🎥 How to provide o1 Pro with your FULL CODEBASE through the ChatGPT <> Cursor connection

yesterday I recorded a video connecting o1 Pro to Cursor through the ChatGPT Desktop app

more than one person commented on how limiting it is for o1 Pro to only have access to a single open file in Cursor

here's a walkthrough of how you can provide o1 Pro with your FULL CODEBASE as context instead of just a single file:

1. Write a Python function that concatenates full codebase into a single snapshot in a .txt file (link in comment)

2. Open that "codebase_snapshot.txt" in a separate Cursor pane

3. Go to the ChatGPT app and ask it what context it has access to through Cursor - should say it has two files with one of them being the codebase snapshot

and boom there ya go. o1 Pro has access to everything in your codebase

full walkthrough here 👇