LabVeira @LabVeira

so far ini 6 fiksi MANTAP yang kubaca tahun ini

🐻 @heninurh

2 days ago

Beli buku apalagi yaaa?? Coba bacaan paling best kalian tahun ini apa guys? Fiksi yaaah. Mau nyontek 😆

70 40 347 39K 259

How to learn Physical Intelligence's robot brain without working at Physical Intelligence: PI was in talks this year to raise $1B at an $11B valuation, and its brain, π0, is open source three of its co-founders teach the ideas inside it on YouTube for $0: Sergey Levine, Chelsea Finn and Quan Vuong here's the order I'd watch them in: 1: copy a human first Levine's Berkeley CS 285, lecture 2 (Spring 2026), is supervised learning of behaviors: a robot copies human demos, then drifts off course after its first small mistake. that drift is why ALOHA and π0 predict a whole chunk of motion at once instead of one step at a time 2: learn the math π0 moves with Stanford AA203, lecture 15 (Spring 2026). an hour in, it explains flow matching, the same math π0 uses to turn a decision into smooth arm motion 3: make it better after the demos run out Chelsea Finn's Stanford CS224R: lecture 16 is RL for real robots, and lecture 17 is a guest lecture from Ashish Kumar, who led AI for Tesla Optimus 4: see the whole brain in one talk Quan Vuong's guest talk in ETH's Robot Learning course is about π0.7. 13 minutes in, it folds a shirt on a robot arm that had no folding data of its own then stop watching and start running it: - openpi, PI's own code and π0 weights, 14.4K stars on GitHub - LeRobot by Hugging Face, 28.5K stars, which fine-tunes π0 on your own demos - a $122 SO-101 arm to record those demos on main insight: the lab in talks at $11 billion is built on three ideas stacked on top of each other: copy a human, move smoothly with flow matching, and let a vision-language model decide what to do next all three are free lectures now. the order is the only part nobody hands you P.S. my Optimus-style robot brain guide below turns this order into an 8-week plan

Martin @0x_meden

4 days ago

Tesla won't sell you Optimus or show you its brain. but every humanoid brain built in public uses the same 3 layers, and all 3 are free to build at home for $0 to $230. last night I got a humanoid waving on my laptop. here's the 8-week path to its brain x.com/i/article/2113…

5 10 66 69K 144

16 buku favoritku sepanjang masa, semua dari 16 ini memiliki alasan personalnya sendiri masing-masing dan sudah ada terjemahan bahasa Indonesia. 1. Le Petit Prince 2. The Count of Monte Cristo 3. Crime and Punishment 4. The Grapes of Wrath

keyboard_arrow_left Previous keyboard_arrow_right Next

If you want to stay up to date on the latest AI and Machine Learning research, reading academic papers can help. But they can be a bit intimidating and hard to understand sometimes. In this course, you'll learn how to approach and understand the theory, math, and structure in academic papers. freecodecamp.org/news/understan…

I spent 4 months collecting this so you don’t have to waste the same time Most people will save this and never open a single link. This is the clean path from zero to being able to build and interview for real AI roles 01 Linear algebra and calculus mml-book.github.io 02 Machine learning foundations youtube.com/playlist?list=… 03 Deep learning from scratch youtube.com/playlist?list=… 04 Neural networks and backprop youtube.com/watch?v=VMj-3S… 05 Computer vision youtube.com/playlist?list=… 06 Natural language processing youtube.com/playlist?list=… 07 Transformers and LLMs cs336.stanford.edu 08 Reinforcement learning rail.eecs.berkeley.edu/deeprlcourse/ 09 Building agents huggingface.co/learn/agents-c… 10 The full path. Free You either walk this now or keep watching random playlists for another year Save this ⭣

shikamaru @shikamarutsai

2 months ago

x.com/i/article/2088…

17 98 439 441K 1K

Tesla's former head of AI put his entire neural network course on YouTube: 10 videos, about 19 hours, and he writes every line of code in front of you, free. "Neural Networks: Zero to Hero," Andrej Karpathy (@karpathy), ex-Sr. Director of AI at Tesla and a founding member of OpenAI. it starts from one neuron and backprop written by hand, then climbs: language models, MLPs, BatchNorm, a WaveNet, GPT built from scratch, the GPT tokenizer and a full GPT-2 reproduction, with all the code on GitHub (24.9K stars). the robot part: π0 is built on PaliGemma, a 3-billion-parameter vision-language model, with a small action expert on top that turns it into motor commands. the transformer inside it is the one he types out in "Let's build GPT." the last video reproduces GPT-2 in 4 hours. from minute 82 to minute 134, he takes one training step from 1,000 ms down to 93 ms with mixed precision, torch.compile and flash attention. Figure is hiring a Helix AI Engineer for training performance right now, at $200K–$400K. the job is to speed up training of 100B+ parameter models on 100,000+ GPUs, and it asks for mixed precision, which is that exact part of the video. Lectures: youtube.com/playlist?list=… Code: github.com/karpathy/nn-ze… P.S. this morning I posted Stanford's CS231N. Karpathy designed the first version of that course and was its primary instructor in 2015. my Optimus-style robot brain guide is below if you want the 8-week path.

Martin @0x_meden

4 days ago

Tesla won't sell you Optimus or show you its brain. but every humanoid brain built in public uses the same 3 layers, and all 3 are free to build at home for $0 to $230. last night I got a humanoid waving on my laptop. here's the 8-week path to its brain x.com/i/article/2113…

5 10 66 69K 144

100% open-source robotic arm! 🦾 @seeedstudio released reBot-DevArm, a robotic arm project lowering the barrier to learning robotics... or how people call it these days: physical AI. Everything is open-sourced. Hardware blueprints include sheet metal and 3D printed parts. Detailed BOM covers every screw with purchase links. Software and algorithms include Python SDK, ROS1/2, Isaac Sim, and LeRobot. Let's have a look at robot specs. So it has 1.5 kg payload, 650 mm max reach, 4.5 kg weight, and less than 0.2 mm repeatability with 6 DoF plus gripper. This is true open source for robotics. When every screw, CAD file, motor driver, and algorithm is freely available, desktop robotic arms become accessible to students, researchers, and developers worldwide. ‼️ Start your career in robotics today: github.com/Seeed-Projects… We need more roboticists than ever! ~~ ♻️ Join the weekly robotics newsletter, and never miss any news → ziegler.substack.com

18 Books = 18 Skills. Thread- 📌🧵

Ternyata olahraga di dalam ruangan dan di luar ruangan punya efek yg beda buat otak. Ada studi dari Nottingham Trent University yg ngetes 45 anak usia 11-13 tahun. Mereka olahraga dengan sesi yg persis sama, sekali di luar ruangan, sekali di dalam ruangan. Sebelum dan sesudahnya, otak mereka dites buat ngukur fokus, kecepatan respon, & kemampuan nahan distraksi. Hasilnya, yg olahraga di luar punya respon yg lebih cepat dan akurasi yg lebih bagus. Efeknya bahkan masih kerasa 45 menit setelah olahraga. Peneliti bilang ini studi pertama yg nunjukin kalau olahraga di luar punya manfaat "kognitif" yg lebih unggul dibanding di dalam ruangan. Jadi kalau lagi mentok belajar, lebih baik kita keluar bentar buat jalan atau lari santai, itu mungkin akan lebih "ngebantu" daripada cuma diem di kamar. Ga harus berat kok. Yg penting keluar dulu aja. Kamu setuju ga sama penelitian ini?

Nicholas Fabiano, MD @NTFabiano

2 days ago

Exercising outside is better for your brain.

31 303 3K 59K 525

x.com/i/article/2077…

books, that'll teach you mathematics

keyboard_arrow_left Previous keyboard_arrow_right Next

x.com/i/article/2114…

We go out of our way to make things more complicated than they need to be. We need to cultivate First Principle thinking is all areas of life. ( 📖 How to Be a Lot Less Stupid, Peter H. Frank , out Nov 1st)

keyboard_arrow_left Previous keyboard_arrow_right Next

📚 spill buku yang saat ini kalian baca

Algumas técnicas de leitura que utilizo pra livros de filosofia: A primeira coisa que normalmente busco fazer é identificar a estrutura do texto e isso quer dizer: 1. Identificar os argumentos e os separar em premissas e conclusões (pressupostos também muitas vezes); 2. Identificar os movimentos de cada argumento. Identificar um movimento é perceber se o autor está encadeando argumentos como uma corrente ou se está apresentando diversos argumentos que não são uma cadeia (nesse caso convém enumerá-los) OU perceber em que momento ele muda a direção da argumentação e passa de uma afirmativa pra uma negativa; Na primeira foto (print do meu caderno) o argumento S. Tomás está decomposto em premissas e conclusões e na foto 3 todo o trabalho está feito dentro do texto em si. Pra isso é importante prestar a atenção nas conjunções, elas indicam reversão de movimento ou continuidade, ou seja, vão negar ou corroborar o que foi dito até aqui (anotações na foto 2). Tudo isso é tido como análise de um texto filosófico e isso é técnica de leitura para se estudar um texto denso. Vi isso em uma disciplina chamada "Leitura e interpretação de textos clássicos". Não faço TUDO isso pra qualquer leitura, somente para aquelas que tenho pretensões acadêmicas e que serão avaliadas (senão leria 2 livros por ano só kkkkkk) A partir daqui eu costumo grifar o argumento principal e as conexões entre ideias que os autores fazem (isso vocês podem ver nos trechos dos meus livros que eu posto). A mera preocupação de grifar o que é mais importante te faz ler com um olhar mais crítico buscando de fato o que é mais importante e isso já transforma sua leitura em leitura ativa, o que é o único modo de ser ler filosofia. A preocupação de notar as conexões entre ideias é para, ao reproduzir a argumentação do autor, você não ficar perdido de como ele faz a passagem de uma conclusão para outras conclusões. É importante conseguir reproduzir todo o movimento que o autor faz. Isso é valiosíssimo em especial pra diálogos Platônicos por exemplo. Eliminar os ruídos de comunicação e se ater à cadeia argumentativa e às objeções. Sobre os comentários/anotações, eu costumo, depois de articular o argumento sob as categorias do autor, transcrever a partir das minhas categorias trazendo outras referências evidenciando contrastes, problemas, soluções ou implicações. No mais, se for uma obra essencial pra sua formação, convém escrever (escrever, não digitar hehe) toda a cadeia argumentativa, tirando tudo que não for essencial e revisitar o seu escrito entre leituras para reforçar o que for essencial. Espero que ajude e conforme for lembrando de mais coisas, vou fazendo mais posts.

keyboard_arrow_left Previous keyboard_arrow_right Next

The scientist who built ImageNet now runs World Labs, and her full Stanford course is on YouTube for free: 18 lectures, about 21 hours, and it ends on robots. "CS231N: Deep Learning for Computer Vision," Fei-Fei Li (@drfeifei), Justin Johnson (@jcjohnss), Ehsan Adeli and Zane Durante, Stanford, Spring 2025. it starts at a linear classifier and climbs: backprop, CNNs, attention and transformers, detection and segmentation, video, self-supervised learning, generative models, 3D vision, vision and language. the robot part: π0 sees the world through SigLIP, a vision transformer. lectures 8 and 16 build that exact piece. lecture 17 is robot learning, taught by Yunzhu Li (@YunzhuLiYZ), whose startup is now part of World Labs too. 62 minutes in, robot arms do kitchen tasks on their own with Diffusion Policy, then a wall of 60 new skills taught the same way. the next slides run from RT-1 to π0, Figure's Helix and NVIDIA's GR00T. Apptronik is hiring a Senior Perception Learning Engineer for its Apollo humanoid right now, at $220K–$250K. the job asks for detection and segmentation, which is lecture 9. Lectures: youtube.com/playlist?list=… Course + assignments: cs231n.stanford.edu/2025/ P.S. a robot needs eyes before it needs a brain. this course is the eyes, and my Optimus-style robot brain guide below is the rest.

Martin @0x_meden

4 days ago

Tesla won't sell you Optimus or show you its brain. but every humanoid brain built in public uses the same 3 layers, and all 3 are free to build at home for $0 to $230. last night I got a humanoid waving on my laptop. here's the 8-week path to its brain x.com/i/article/2113…

5 10 66 69K 144

Untuk teman-teman yang ingin mendaftar sebagai anggota Perpus Berjalan simak thread ini yuk.

“Knowledge isn't free. You have to pay attention.” — Richard Feynman

Before you build a robot, watch this. Here are 18 open source robots, shown with their 3D models. 1. Microban 2. MiniLoong 3. Poppy Humanoid 4. Otto DIY 5. pib 6. Mini Pupper 7. Petoi Bittle 8. Stanford Doggo 9. SpotMicro 10. OpenQuadruped 11. Puguette 12. SO-ARM100 13. RoArm-M2 14. OpenMANIPULATOR-X 15. LeKiwi 16. TurtleBot3 Burger 17. OpenBot 18. Duckiebot These projects cover humanoids, quadrupeds, robotic arms and mobile robots. Before building, check each project's official repository. Hardware specifications, 3D design files and licences vary between projects. My personal pick would be SO-ARM100. I'd like to explore its design and see what I could build with it. Which of these 18 robots would you build first? 3D model credits: Microban (Rhoban, CC BY-NC-SA 4.0) · Poppy Humanoid (Poppy Project, CC BY-SA 4.0) · Otto DIY (ottobot. org, CC BY-SA 4.0) · SpotMicro design by Deok-yeon Kim / KDY0523 (CC BY 3.0) · OpenBot wheel model: FreeCAD library (CC BY 3.0) · Duckiebot model: Duckietown (duckietown. com) · All other models from their respective project repositories.

Techniahqrobot | humanoid robots @techniahqrobot

3 days ago

Most people miss this one detail about robot datasets: failure is data too. AGIBOT WORLD 2026 Release 4 focuses on physical reasoning. The robot must track objects handle occlusion respect space constraints and decide what to do next. The dataset covers 9 reasoning abilities

1 1 14 9K 12