Загрузка видео...

Не удалось загрузить видео

Возникла проблема при загрузке этого видео. Это может быть связано с временными проблемами сети или видео может быть недоступно.

На главную

🎯 The Mechanics Training Map (V1) 🗺️ CODE: 8458-1150-5480 -16+ Aim & Kovaaks Scenarios -20+ Piece Control Scenarios -20+ Edit Scenarios -Competitive Training Section -Aim Score, Accuracy & Timer Tracking -Freebuild Training -Much, much more... Everything you need to improve at Fortnite in ONE map. More features & updates... show more

Weaside

11,518 subscribers

10,700 просмотров • 1 месяц назад •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

TRAINING HUB V2 -Custom Routine Creator -20 Piece Control Scenarios with Perfect Peek option -Smart improve System -Competitive mechanics section -Seasonal updates -Much, much more Me and unixo maps cooked, I'll be following back some people who retweet this and favorite the map!

TRAINING HUB V2 -Custom Routine Creator -20 Piece Control Scenarios with Perfect Peek option -Smart improve System -Competitive mechanics section -Seasonal updates -Much, much more Me and unixo maps cooked, I'll be following back some people who retweet this and favorite the map!

Reisshub

152,642 просмотров • 6 месяцев назад

THE TRAINING HUB (5257-9815-3361) The practice map designed for competitive players! This map has everything you need to practice as a competitive player: -Bullet drop off trainer -6 different -Edit crosshair training -Piece control scenarios with perfect peeking settings -Adjustable speeds map wide -And tonnes tonnes more. Me and unixo maps are really proud of this one, I'll be following back some people who retweet this and favorite the map!

THE TRAINING HUB (5257-9815-3361) The practice map designed for competitive players! This map has everything you need to practice as a competitive player: -Bullet drop off trainer -6 different -Edit crosshair training -Piece control scenarios with perfect peeking settings -Adjustable speeds map wide -And tonnes tonnes more. Me and unixo maps are really proud of this one, I'll be following back some people who retweet this and favorite the map!

Reisshub

338,444 просмотров • 2 лет назад

I beg some of these people to take some time in the training area, just 10-20 minutes tops, to learn the basics of the mechanics. That clip just reinforces why devs add a training mode to games like this, why it's in your best interest to at least practice basic mechanics and combos. It's crazy how much more fun a game can be when you at least try to engage with what the game gives you instead of doing the bare minimum.

I beg some of these people to take some time in the training area, just 10-20 minutes tops, to learn the basics of the mechanics. That clip just reinforces why devs add a training mode to games like this, why it's in your best interest to at least practice basic mechanics and combos. It's crazy how much more fun a game can be when you at least try to engage with what the game gives you instead of doing the bare minimum.

Limitless

21,173 просмотров • 1 год назад

I've been playing more with this depth map motion transfer workflow, and if you're trying to recreate really specific body movements from a reference video, this works surprisingly well. Here's the workflow: 1. Grab your reference video, then create your character and environment references. 2. Run the video through Depth Anything on fal to generate a depth map: 3. Drop that depth map into Seedance Omni as your motion reference along with your character and environment references. You don't even need the original video. I also tested this against using the original video as the motion reference in Seedance, and there's definitely a noticeable difference. One thing I found, though, is that this works much better for body movement than facial performance since the depth map can't really capture subtle facial motion. I'm running more tests to see if there's a good way to control both body movement and facial performance. If you have any workflows I should try, do share in the comments!

I've been playing more with this depth map motion transfer workflow, and if you're trying to recreate really specific body movements from a reference video, this works surprisingly well. Here's the workflow: 1. Grab your reference video, then create your character and environment references. 2. Run the video through Depth Anything on fal to generate a depth map: 3. Drop that depth map into Seedance Omni as your motion reference along with your character and environment references. You don't even need the original video. I also tested this against using the original video as the motion reference in Seedance, and there's definitely a noticeable difference. One thing I found, though, is that this works much better for body movement than facial performance since the depth map can't really capture subtle facial motion. I'm running more tests to see if there's a good way to control both body movement and facial performance. If you have any workflows I should try, do share in the comments!

Curious Refuge

16,487 просмотров • 1 месяц назад

Asked God’s Eye View to explain why there’s so much military activity in this area -- all using voice, while it sees what I’m seeing on the 3D map. Turns out Fort Rucker is a hot spot for helicopter training. So I jumped into cockpit view to track one of the choppers live in the middle of a drill. IMO gpt-realtime-2 is quite underrated for geospatial apps. Voice is such a natural interface for exploring the world and the data underpinning it. Code is open source if you wanna try it.

Asked God’s Eye View to explain why there’s so much military activity in this area -- all using voice, while it sees what I’m seeing on the 3D map. Turns out Fort Rucker is a hot spot for helicopter training. So I jumped into cockpit view to track one of the choppers live in the middle of a drill. IMO gpt-realtime-2 is quite underrated for geospatial apps. Voice is such a natural interface for exploring the world and the data underpinning it. Code is open source if you wanna try it.

Bilawal Sidhu

30,721 просмотров • 24 дней назад

How is that boring Nick Eh 30 ??? Coming from you, I think that's just bad faith. Given the amount of work we put into this map & the time you've spent "playing" it, your opinion is clearly not objective. There are literally a lot animated quests, a battle pass with seasons and lots of cosmetics to use inside... in short, everything to let players having fun & coming back, and this has been proven because it's been the most played Impostor map in discovery for over a year! When you're playing the map with over 5k viewers just to criticise our creation without even staying a whole round ; it's just strange. So I don't know if you had a reason to be like that on our map, but seeing you having fun later on maps that are much more boring (imo), I don't know how to take it...

How is that boring Nick Eh 30 ??? Coming from you, I think that's just bad faith. Given the amount of work we put into this map & the time you've spent "playing" it, your opinion is clearly not objective. There are literally a lot animated quests, a battle pass with seasons and lots of cosmetics to use inside... in short, everything to let players having fun & coming back, and this has been proven because it's been the most played Impostor map in discovery for over a year! When you're playing the map with over 5k viewers just to criticise our creation without even staying a whole round ; it's just strange. So I don't know if you had a reason to be like that on our map, but seeing you having fun later on maps that are much more boring (imo), I don't know how to take it...

Volviik 👑

312,606 просмотров • 2 лет назад

Cache is back in CS2, and this is exactly how you make a Counter-Strike map. They completely remade the map from the ground up, absolutely nothing remains from the version from the workshop. Gameplay on the new cache is very clean and smooth without too much physical or visual clutter. Many of the more complicated props have been replaced with simple boxes, such as the box on forklift in A site. There are a few clipping issues with props and walls, but I imagine these will get sorted out over time. The visual fidelity is amazing without sacrificing performance. This will be a very easy map to learn and play, as there isn't very much complexity at all. Many of the features we all kinda hated are gone, such as the window at Z connector mid, new box, and the railing at highway. If they remove a map for this, it should be a simple 4 square map, otherwise the map pool is going to get too filled with easy maps. Remember when we got like... NO updates. Now I'm getting exhausted covering all of this week after week. THANK GOODNESS.

Cache is back in CS2, and this is exactly how you make a Counter-Strike map. They completely remade the map from the ground up, absolutely nothing remains from the version from the workshop. Gameplay on the new cache is very clean and smooth without too much physical or visual clutter. Many of the more complicated props have been replaced with simple boxes, such as the box on forklift in A site. There are a few clipping issues with props and walls, but I imagine these will get sorted out over time. The visual fidelity is amazing without sacrificing performance. This will be a very easy map to learn and play, as there isn't very much complexity at all. Many of the features we all kinda hated are gone, such as the window at Z connector mid, new box, and the railing at highway. If they remove a map for this, it should be a simple 4 square map, otherwise the map pool is going to get too filled with easy maps. Remember when we got like... NO updates. Now I'm getting exhausted covering all of this week after week. THANK GOODNESS.

WarOwl

366,239 просмотров • 4 месяцев назад

New Course: Post-training of LLMs Learn to post-train and customize an LLM in this short course, taught by Banghua Zhu, Assistant Professor at the University of Washington University of Washington, and co-founder of @NexusflowX. Training an LLM to follow instructions or answer questions has two key stages: pre-training and post-training. In pre-training, it learns to predict the next word or token from large amounts of unlabeled text. In post-training, it learns useful behaviors such as following instructions, tool use, and reasoning. Post-training transforms a general-purpose token predictor—trained on trillions of unlabeled text tokens—into an assistant that follows instructions and performs specific tasks. Because it is much cheaper than pre-training, it is practical for many more teams to incorporate post-training methods into their workflows than pre-training. In this course, you’ll learn three common post-training methods—Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Online Reinforcement Learning (RL)—and how to use each one effectively. With SFT, you train the model on pairs of input and ideal output responses. With DPO, you provide both a preferred (chosen) and a less preferred (rejected) response and train the model to favor the preferred output. With RL, the model generates an output, receives a reward score based on human or automated feedback, and updates the model to improve performance. You’ll learn the basic concepts, common use cases, and principles for curating high-quality data for effective training. Through hands-on labs, you’ll download a pre-trained model from Hugging Face and post-train it using SFT, DPO, and RL to see how each technique shapes model behavior. In detail, you’ll: - Understand what post-training is, when to use it, and how it differs from pre-training. - Build an SFT pipeline to turn a base model into an instruct model. - Explore how DPO reshapes behavior by minimizing contrastive loss—penalizing poor responses and reinforcing preferred ones. - Implement a DPO pipeline to change the identity of a chat assistant. - Learn online RL methods such as Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO), and how to design reward functions. - Train a model with GRPO to improve its math capabilities using a verifiable reward. Post-training is one of the most rapidly developing areas of LLM training. Whether you’re building a high-accuracy context-specific assistant, fine-tuning a model's tone, or improving task-specific accuracy, this course will give you experience with the most important techniques shaping how LLMs are post-trained today. Please sign up here:

New Course: Post-training of LLMs Learn to post-train and customize an LLM in this short course, taught by Banghua Zhu, Assistant Professor at the University of Washington University of Washington, and co-founder of @NexusflowX. Training an LLM to follow instructions or answer questions has two key stages: pre-training and post-training. In pre-training, it learns to predict the next word or token from large amounts of unlabeled text. In post-training, it learns useful behaviors such as following instructions, tool use, and reasoning. Post-training transforms a general-purpose token predictor—trained on trillions of unlabeled text tokens—into an assistant that follows instructions and performs specific tasks. Because it is much cheaper than pre-training, it is practical for many more teams to incorporate post-training methods into their workflows than pre-training. In this course, you’ll learn three common post-training methods—Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Online Reinforcement Learning (RL)—and how to use each one effectively. With SFT, you train the model on pairs of input and ideal output responses. With DPO, you provide both a preferred (chosen) and a less preferred (rejected) response and train the model to favor the preferred output. With RL, the model generates an output, receives a reward score based on human or automated feedback, and updates the model to improve performance. You’ll learn the basic concepts, common use cases, and principles for curating high-quality data for effective training. Through hands-on labs, you’ll download a pre-trained model from Hugging Face and post-train it using SFT, DPO, and RL to see how each technique shapes model behavior. In detail, you’ll: - Understand what post-training is, when to use it, and how it differs from pre-training. - Build an SFT pipeline to turn a base model into an instruct model. - Explore how DPO reshapes behavior by minimizing contrastive loss—penalizing poor responses and reinforcing preferred ones. - Implement a DPO pipeline to change the identity of a chat assistant. - Learn online RL methods such as Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO), and how to design reward functions. - Train a model with GRPO to improve its math capabilities using a verifiable reward. Post-training is one of the most rapidly developing areas of LLM training. Whether you’re building a high-accuracy context-specific assistant, fine-tuning a model's tone, or improving task-specific accuracy, this course will give you experience with the most important techniques shaping how LLMs are post-trained today. Please sign up here:

Andrew Ng

125,146 просмотров • 1 год назад

The fruit fly's brain is superior to the human brain, and GPT-6 ASTRA has proven this. And the internet went crazy. Most AI models learn by training on massive datasets, running billions of computations to acquire new skills. But researchers just demonstrated a radically different approach: No giant dataset. No traditional training. Just take the physical wiring of a biological brain and simulate it in a computer. The digital fruit fly reached ~91% behavioral accuracy using only 4 things: The complete neural connectivity map Synaptic connection counts as connection weights Excitatory vs inhibitory neuron types A simple Leaky Integrate-and-Fire model The wild part? The wiring itself may contain much more of the brain’s “code” than we thought. From tens of thousands of neurons in a fly → millions in a mouse → ~86 billion in a human. If this scales, whole-brain simulation could become one of the biggest frontiers in AI. The era of biologically inspired intelligence may have just begun

The fruit fly's brain is superior to the human brain, and GPT-6 ASTRA has proven this. And the internet went crazy. Most AI models learn by training on massive datasets, running billions of computations to acquire new skills. But researchers just demonstrated a radically different approach: No giant dataset. No traditional training. Just take the physical wiring of a biological brain and simulate it in a computer. The digital fruit fly reached ~91% behavioral accuracy using only 4 things: The complete neural connectivity map Synaptic connection counts as connection weights Excitatory vs inhibitory neuron types A simple Leaky Integrate-and-Fire model The wild part? The wiring itself may contain much more of the brain’s “code” than we thought. From tens of thousands of neurons in a fly → millions in a mouse → ~86 billion in a human. If this scales, whole-brain simulation could become one of the biggest frontiers in AI. The era of biologically inspired intelligence may have just begun

Zentrix⌚️

101,277 просмотров • 6 дней назад

If you're looking for a weekend project, how about training your own text-to-speech model from scratch on your own GPU, and then running it on any device's CPU? We just open-sourced the entire Pocket TTS training stack: data pipeline, recipes, and evals. It learns pretty damn fast: ~15k steps: babbling starts turning into words ~50k steps: it reads anything you type (WER under 1%) ~200k steps: the voice stops sounding synthetic On a beefy consumer GPU, that's a week of training. On eight H100s: 10-20 hours. A TTS training run will cost you less than $200 if you rent your hardware, and an order of magnitude less if you just pay for power. Some things we'd love to see people try: - Train it in your own language (a few hundred hours of speech gets you surprisingly far). - Add new features to Pocket TTS (Emotion tags? Make it sing?). - Beat us at our own game: make it faster and smaller. Show us what you build! We'll highlight the best models and new languages for the whole community to enjoy. Pocket TTS has already found many use cases, from reading for people with visual impairments to making NPCs in video games talk, and we're sure there's much more to do with it! Here's an example of a Czech Pocket TTS. Try just asking your favorite agent to find data and apply the method, and you can have your own. Get started:

If you're looking for a weekend project, how about training your own text-to-speech model from scratch on your own GPU, and then running it on any device's CPU? We just open-sourced the entire Pocket TTS training stack: data pipeline, recipes, and evals. It learns pretty damn fast: ~15k steps: babbling starts turning into words ~50k steps: it reads anything you type (WER under 1%) ~200k steps: the voice stops sounding synthetic On a beefy consumer GPU, that's a week of training. On eight H100s: 10-20 hours. A TTS training run will cost you less than $200 if you rent your hardware, and an order of magnitude less if you just pay for power. Some things we'd love to see people try: - Train it in your own language (a few hundred hours of speech gets you surprisingly far). - Add new features to Pocket TTS (Emotion tags? Make it sing?). - Beat us at our own game: make it faster and smaller. Show us what you build! We'll highlight the best models and new languages for the whole community to enjoy. Pocket TTS has already found many use cases, from reading for people with visual impairments to making NPCs in video games talk, and we're sure there's much more to do with it! Here's an example of a Czech Pocket TTS. Try just asking your favorite agent to find data and apply the method, and you can have your own. Get started:

kyutai

105,352 просмотров • 24 дней назад

How to Create AI Videos with Google Flow from Start to Finish. Google Flow can do much more than generate short AI clips. It gives you everything you need to plan, create, edit, and refine AI videos in one workflow. In this tutorial, you'll learn how to build AI videos from scratch, create consistent characters, generate cinematic scenes, edit your projects, and improve existing videos using Google Flow. Whether you're creating YouTube videos, short films, or social media content, this guide shows you the core features and practical workflow you can apply to your own projects. Like, Retweet, Bookmark & Comment💯 Follow BIG CHRIS 🤝 Video Credit: Kevin Stratvert (

How to Create AI Videos with Google Flow from Start to Finish. Google Flow can do much more than generate short AI clips. It gives you everything you need to plan, create, edit, and refine AI videos in one workflow. In this tutorial, you'll learn how to build AI videos from scratch, create consistent characters, generate cinematic scenes, edit your projects, and improve existing videos using Google Flow. Whether you're creating YouTube videos, short films, or social media content, this guide shows you the core features and practical workflow you can apply to your own projects. Like, Retweet, Bookmark & Comment💯 Follow BIG CHRIS 🤝 Video Credit: Kevin Stratvert (

BIG CHRIS

24,147 просмотров • 2 месяцев назад

Today at Yoast we’re shipping something I've actually wanted to see on the web for some time (even before working at Yoast) and am now honored to be part of the team to bring this to the masses... Built in collaboration with the Open Source and NLWeb team at Microsoft lead by R.V.Guha (co-creator of RSS, RDF and other web standards) - we're introducing the Schema Aggregation feature: a "schemamap" endpoint that outputs your site's entire structured data map in one place. Under the hood, we now provide a standardised, deduplicated map of your entities via a single endpoint. An agent no longer needs to crawl all individual pages to understand its meaning but can now ingest an entire entity map with ease. A few details to note about the endpoint: - It's is cacheable with sub‑100ms responses - It respects existing privacy and indexing settings - It aggregates all indexable content without navigation noise - It merges duplicate entities so your "Author X" or "Article Y" exist as a single node instead of being re‑discovered on every URL. - If you're using one of our paid plugins that extends schema even more (such as Yoast WooCommerce SEO adds product schema) this will be populated within the endpoint too - If you already extend Yoast’s Schema API, or use partners like events or recipe plugins, their entities are pulled into the same map automatically. For me, this is one of the first major ways the agentic web can ingest a site at scale, and with much more efficiency and context. It has also been so much fun to work with this alongside the team at Yoast, particularly the genius mind of Alain Schlesser. Lastly, we have also launched a schema visualisation tool to view how everything is output. You can enable this feature today with the free version of Yoast SEO.

Today at Yoast we’re shipping something I've actually wanted to see on the web for some time (even before working at Yoast) and am now honored to be part of the team to bring this to the masses... Built in collaboration with the Open Source and NLWeb team at Microsoft lead by R.V.Guha (co-creator of RSS, RDF and other web standards) - we're introducing the Schema Aggregation feature: a "schemamap" endpoint that outputs your site's entire structured data map in one place. Under the hood, we now provide a standardised, deduplicated map of your entities via a single endpoint. An agent no longer needs to crawl all individual pages to understand its meaning but can now ingest an entire entity map with ease. A few details to note about the endpoint: - It's is cacheable with sub‑100ms responses - It respects existing privacy and indexing settings - It aggregates all indexable content without navigation noise - It merges duplicate entities so your "Author X" or "Article Y" exist as a single node instead of being re‑discovered on every URL. - If you're using one of our paid plugins that extends schema even more (such as Yoast WooCommerce SEO adds product schema) this will be populated within the endpoint too - If you already extend Yoast’s Schema API, or use partners like events or recipe plugins, their entities are pulled into the same map automatically. For me, this is one of the first major ways the agentic web can ingest a site at scale, and with much more efficiency and context. It has also been so much fun to work with this alongside the team at Yoast, particularly the genius mind of Alain Schlesser. Lastly, we have also launched a schema visualisation tool to view how everything is output. You can enable this feature today with the free version of Yoast SEO.

Alex Moss

15,414 просмотров • 6 месяцев назад

Here's a sneaky peak the next release of the GB Renewables Map, or should I say the GB Not Just a Renewables Map… That's right, all types of power assets are coming to the map! Plus a whole bunch of other improvements. It's been an absolutely massive piece of work that's sucked up most of my evenings and free time, but I'm very nearly there now. And if you didn't already know, yup, this a hobby project I do for fun outside of my day job… because I'm weird like that. This video shows how things currently look with all the power assets visible, though you might also notice a few other additions if you look closely… It still has some rough edges so I'm in two minds whether to just put this live and work on fixes and improvements as I go. This is a huge change to the code and to the whole concept for the map, so I'm not entirely sure I'll ever have it 100% nailed down. I'm also cautious letting perfect be the enemy of good enough. I don't think there's anything awfully broken though, I've tested it so much it's probably more robust than the current version of the map! Though there will inevitably be things that I've overlooked. Even in its current state it's already proven incredibly valuable for me to learn more about what's happening during periods of constraint on the grid, like seeing which assets are being turned up to balance things out. What do you think? Shall I put this live and we can find the quirks together?

Here's a sneaky peak the next release of the GB Renewables Map, or should I say the GB Not Just a Renewables Map… That's right, all types of power assets are coming to the map! Plus a whole bunch of other improvements. It's been an absolutely massive piece of work that's sucked up most of my evenings and free time, but I'm very nearly there now. And if you didn't already know, yup, this a hobby project I do for fun outside of my day job… because I'm weird like that. This video shows how things currently look with all the power assets visible, though you might also notice a few other additions if you look closely… It still has some rough edges so I'm in two minds whether to just put this live and work on fixes and improvements as I go. This is a huge change to the code and to the whole concept for the map, so I'm not entirely sure I'll ever have it 100% nailed down. I'm also cautious letting perfect be the enemy of good enough. I don't think there's anything awfully broken though, I've tested it so much it's probably more robust than the current version of the map! Though there will inevitably be things that I've overlooked. Even in its current state it's already proven incredibly valuable for me to learn more about what's happening during periods of constraint on the grid, like seeing which assets are being turned up to balance things out. What do you think? Shall I put this live and we can find the quirks together?

Robin Hawkes

19,444 просмотров • 7 месяцев назад

Military Emergency Manager Has A Warning ⛔️ Just In Time For Elections FEMA Has Put Out New Risks For 2024 - Infrastructure Blackouts - Social Unrest - Mass Casualty/ Mass Shooter Specifically Focused Around Loss Of Life How does FEMA know this is about to happen? “Some information that I just want to pass along, give to you guys about FEMA because a lot of people don't understand how FEMA operates. So FEMA provides training for things that they see on the near future. They use predictive modeling, threat assessment, um, following trends to look at possible scenarios that might coming up. This year, the first 3 scenarios that they're offering to emergency managers like myself in these virtual tabletop exercises, is infrastructure blackouts, social unrest, And mass casualty, mass shooter instance specifically focused around loss. So given that information, I would highly recommend you to think about your planning processes, your systems, And how you're just gonna conduct yourself in terms of one of those 3 or more not occurring in your daily life. Something to think about, something to keep in mind when you see FEMA putting information out. Hope this helps”

Military Emergency Manager Has A Warning ⛔️ Just In Time For Elections FEMA Has Put Out New Risks For 2024 - Infrastructure Blackouts - Social Unrest - Mass Casualty/ Mass Shooter Specifically Focused Around Loss Of Life How does FEMA know this is about to happen? “Some information that I just want to pass along, give to you guys about FEMA because a lot of people don't understand how FEMA operates. So FEMA provides training for things that they see on the near future. They use predictive modeling, threat assessment, um, following trends to look at possible scenarios that might coming up. This year, the first 3 scenarios that they're offering to emergency managers like myself in these virtual tabletop exercises, is infrastructure blackouts, social unrest, And mass casualty, mass shooter instance specifically focused around loss. So given that information, I would highly recommend you to think about your planning processes, your systems, And how you're just gonna conduct yourself in terms of one of those 3 or more not occurring in your daily life. Something to think about, something to keep in mind when you see FEMA putting information out. Hope this helps”

Wall Street Apes

153,904 просмотров • 2 лет назад

I talked to Eoghan McCabe about why you should be posting more: "Personally, you see every piece of content you post. And if you post a lot, it might feel like you're posting too much. But because of algorithmic feeds, no one else sees everything you post. So you can't assume everyone sees every single post. The average person will only see 10%, 20@, maybe 50% of what you post. No one will actually see all of it. And you don't know what's gonna work when you're posting. So you really just need to put a lot of shots on goal. It's like having a high shipping velocity on product. Imagine it takes 100 posts for something to break through. If you post 10x a day, it will take 10 days. If you post once a day, it will take over a quarter. Once a year, it will take 100 years. This is an advantage that some younger founders have. People like Amjad at Replit, he can work and tweet. He's built a great product. And he's constantly promoting it."

I talked to Eoghan McCabe about why you should be posting more: "Personally, you see every piece of content you post. And if you post a lot, it might feel like you're posting too much. But because of algorithmic feeds, no one else sees everything you post. So you can't assume everyone sees every single post. The average person will only see 10%, 20@, maybe 50% of what you post. No one will actually see all of it. And you don't know what's gonna work when you're posting. So you really just need to put a lot of shots on goal. It's like having a high shipping velocity on product. Imagine it takes 100 posts for something to break through. If you post 10x a day, it will take 10 days. If you post once a day, it will take over a quarter. Once a year, it will take 100 years. This is an advantage that some younger founders have. People like Amjad at Replit, he can work and tweet. He's built a great product. And he's constantly promoting it."

The Peel

189,314 просмотров • 9 месяцев назад

After testing FSD v14 head-to-head against other top-tier intelligent driving systems, this Chinese defensive-driving instructor shared his takeaway: “After testing FSD, my conclusion is: this is the best overall experience among all the driver-assist systems I’ve tested. As someone who’s been a defensive-driving instructor for almost 20 years, I can say with full confidence that its anticipation and decision-making in the vast majority of complex road conditions is pretty much in line with my own driving habits — some of it is even identical. Throughout the whole process, my anxiety around needing to take over was extremely low. The occasional interventions and takeovers I did do were mostly because it wasn’t familiar with China’s laws, regulations, signs and markings, or with certain special road conditions. And some were takeovers just for the sake of efficiency. Based on my nearly 20 years of training and evaluating all kinds of professional and non-professional drivers, I can say that in the vast majority of scenarios, FSD’s decision quality and decision stability have already surpassed most experienced human drivers. And one more thing to note: it has no emotions. It never gets tired. It never gets distracted. And an algorithm that hasn’t even been specifically trained on Chinese road conditions yet — relying only on its underlying general-purpose risk perception and avoidance logic — has already managed, in most scenarios, to drive defensively like a veteran driver. This is a genuinely remarkable display of true end-to-end capability. Because true end-to-end learning isn’t about rules. It’s about the physical laws of the real world and the physiological patterns of humans. And this makes me really look forward to seeing how the official version of FSD performs once local training is complete. But I do have some concerns too. I’m also worried: once FSD starts training in China, whether it might pick up some of the bad driving habits of Chinese drivers. This mainly depends on Tesla’s data cleaning and filtering work. We’ll have to wait and see.”

After testing FSD v14 head-to-head against other top-tier intelligent driving systems, this Chinese defensive-driving instructor shared his takeaway: “After testing FSD, my conclusion is: this is the best overall experience among all the driver-assist systems I’ve tested. As someone who’s been a defensive-driving instructor for almost 20 years, I can say with full confidence that its anticipation and decision-making in the vast majority of complex road conditions is pretty much in line with my own driving habits — some of it is even identical. Throughout the whole process, my anxiety around needing to take over was extremely low. The occasional interventions and takeovers I did do were mostly because it wasn’t familiar with China’s laws, regulations, signs and markings, or with certain special road conditions. And some were takeovers just for the sake of efficiency. Based on my nearly 20 years of training and evaluating all kinds of professional and non-professional drivers, I can say that in the vast majority of scenarios, FSD’s decision quality and decision stability have already surpassed most experienced human drivers. And one more thing to note: it has no emotions. It never gets tired. It never gets distracted. And an algorithm that hasn’t even been specifically trained on Chinese road conditions yet — relying only on its underlying general-purpose risk perception and avoidance logic — has already managed, in most scenarios, to drive defensively like a veteran driver. This is a genuinely remarkable display of true end-to-end capability. Because true end-to-end learning isn’t about rules. It’s about the physical laws of the real world and the physiological patterns of humans. And this makes me really look forward to seeing how the official version of FSD performs once local training is complete. But I do have some concerns too. I’m also worried: once FSD starts training in China, whether it might pick up some of the bad driving habits of Chinese drivers. This mainly depends on Tesla’s data cleaning and filtering work. We’ll have to wait and see.”

Ray

28,211 просмотров • 11 дней назад

Nothing beats open-source! We now have a fully open-source coding agent that can compete with any proprietary solution out there: • 500+ available models • It doesn't train on your code • It doesn't charge a penny for tokens • It's open. You can change if you want. Kilo Code is the #1 app on OpenRouter right now, and the community is releasing new features on GitHub at rocket speed (about 20 releases in the last 30 days). I installed it in Visual Studio Code, but you can install it in pretty much any IDE. I need to spend more time with it, but here is a very interesting design choice: Most coding agents support two modes: an "agentic" mode for writing code and an "ask" mode for talking to the model without making any changes. Kilo Code supports 5 different modes: 1. Architect mode - Research, design, and planning 2. Debug mode - Problem solving and troubleshooting 3. Code mode - To write code. 4. Ask mode - To talk to the model. 5. Orchestrator mode - Combine every agent to solve problems. On paper, I really like this. I'll need more time to try out the specialized agents and see whether niching becomes an advantage. Here is the link to the website: Thanks to the Kilo for partnering with me on this post.

Nothing beats open-source! We now have a fully open-source coding agent that can compete with any proprietary solution out there: • 500+ available models • It doesn't train on your code • It doesn't charge a penny for tokens • It's open. You can change if you want. Kilo Code is the #1 app on OpenRouter right now, and the community is releasing new features on GitHub at rocket speed (about 20 releases in the last 30 days). I installed it in Visual Studio Code, but you can install it in pretty much any IDE. I need to spend more time with it, but here is a very interesting design choice: Most coding agents support two modes: an "agentic" mode for writing code and an "ask" mode for talking to the model without making any changes. Kilo Code supports 5 different modes: 1. Architect mode - Research, design, and planning 2. Debug mode - Problem solving and troubleshooting 3. Code mode - To write code. 4. Ask mode - To talk to the model. 5. Orchestrator mode - Combine every agent to solve problems. On paper, I really like this. I'll need more time to try out the specialized agents and see whether niching becomes an advantage. Here is the link to the website: Thanks to the Kilo for partnering with me on this post.

Santiago

73,062 просмотров • 10 месяцев назад

In my previous post, I shared my first impressions after trying BingX Ultra TradingView. But for me, a trading feature only proves its value when it actually improves my workflow in real market conditions not just on paper. So over the past few days, I decided to trade exclusively with Ultra TradingView and record the entire process to see whether it could genuinely make mobile trading more efficient. 1⃣ Workflow #1: Managing an Open BTC Position After identifying a support zone, I opened a BTC long at 59.397. 🎯 Take Profit: 59.491 🛡 Stop Loss: 59.342 What I really wanted to test wasn't the order execution speed, but how convenient it is to manage an open position afterward. Instead of switching to a separate Position panel like I usually do on other platforms, I simply dragged both my Take Profit and Stop Loss levels directly on the chart. Every adjustment was reflected instantly, making the entire position management process much more intuitive. 2⃣ Workflow #2: Planning Two ETH Scenarios For ETH, I didn't rush into a market order. Instead, I prepared two trading scenarios in advance. 📍 Long Entry: 1,564 📍 Short Entry: 1,573 Both pending orders were placed directly within the TradingView chart, allowing me to monitor my trading plan, adjust entry levels when needed, and manage multiple scenarios without leaving the chart. After several days of using Ultra TradingView, what impressed me most wasn't the 108 technical indicators or 86 drawing tools. It was the trading workflow itself. From market analysis to order placement and position management, everything happens on a single chart. For someone who trades frequently on mobile, eliminating unnecessary screen switching makes a noticeable difference especially during fast-moving markets. I've included the full trading session in the video below so you can see exactly how I used Ultra TradingView across these two different trading scenarios. If you've already tried it, I'd love to hear how it fits into your own trading workflow. #BingXUltraTV BingX

In my previous post, I shared my first impressions after trying BingX Ultra TradingView. But for me, a trading feature only proves its value when it actually improves my workflow in real market conditions not just on paper. So over the past few days, I decided to trade exclusively with Ultra TradingView and record the entire process to see whether it could genuinely make mobile trading more efficient. 1⃣ Workflow #1: Managing an Open BTC Position After identifying a support zone, I opened a BTC long at 59.397. 🎯 Take Profit: 59.491 🛡 Stop Loss: 59.342 What I really wanted to test wasn't the order execution speed, but how convenient it is to manage an open position afterward. Instead of switching to a separate Position panel like I usually do on other platforms, I simply dragged both my Take Profit and Stop Loss levels directly on the chart. Every adjustment was reflected instantly, making the entire position management process much more intuitive. 2⃣ Workflow #2: Planning Two ETH Scenarios For ETH, I didn't rush into a market order. Instead, I prepared two trading scenarios in advance. 📍 Long Entry: 1,564 📍 Short Entry: 1,573 Both pending orders were placed directly within the TradingView chart, allowing me to monitor my trading plan, adjust entry levels when needed, and manage multiple scenarios without leaving the chart. After several days of using Ultra TradingView, what impressed me most wasn't the 108 technical indicators or 86 drawing tools. It was the trading workflow itself. From market analysis to order placement and position management, everything happens on a single chart. For someone who trades frequently on mobile, eliminating unnecessary screen switching makes a noticeable difference especially during fast-moving markets. I've included the full trading session in the video below so you can see exactly how I used Ultra TradingView across these two different trading scenarios. If you've already tried it, I'd love to hear how it fits into your own trading workflow. #BingXUltraTV BingX

DGroup Capital 🇻🇳

19,651 просмотров • 2 месяцев назад