正在加载视频...

视频加载失败

返回首页

🧱We've heard your feedback - some updates before the weekend: 1. Hot section - most improvement over the past hour, shows trends in real time 2. Semantic search provided by Mixedbread - search w/ natural language `npx skills` updates coming soon ;)

andrewqu's profile picture

Andrew Qu

3,259 subscribers

38,560 次观看 • 6 个月前 •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

New year, new features 🚀 This past month, our YouTube teams introduced: 🎶 Several updates to our YouTubemusic app: like new ways to create playlists from prompts, filter your “Liked Music” playlist by genre and mood filter chips, and search inside your playlists ⚡Ingredients to Video in Shorts and the YouTube Create app, which allows you to create high quality, vertical video clips with up to three photos 📍A new option for you to enable your device location to be used, so you can see nearby search results and recommendations 🎛️ Search filters for Shorts and long-form videos 🌍 A GeoGuessr playable game More coming soon!
0:11

New year, new features 🚀 This past month, our YouTube teams introduced: 🎶 Several updates to our YouTubemusic app: like new ways to create playlists from prompts, filter your “Liked Music” playlist by genre and mood filter chips, and search inside your playlists ⚡Ingredients to Video in Shorts and the YouTube Create app, which allows you to create high quality, vertical video clips with up to three photos 📍A new option for you to enable your device location to be used, so you can see nearby search results and recommendations 🎛️ Search filters for Shorts and long-form videos 🌍 A GeoGuessr playable game More coming soon!

Neal Mohan

14,522 次观看 • 5 个月前

🪖 $40K in trader incentives sponsored by Bluefin. Rewards will be based on platform interactions, leaderboard + full details coming soon. In the past 48 hours, we’ve shipped major updates: • Mobile View • PNL Cards • Improved bonding progress status update • Portfolio Page • First Dev Buy with Sniper Protection • User Trade Stats • Improved token discovery & visibility • Search Bar (CA or Name) • Enhanced X link embeds And we’re just getting started, dozens more features & quality-of-life upgrades are on the way.
0:15

🪖 $40K in trader incentives sponsored by Bluefin. Rewards will be based on platform interactions, leaderboard + full details coming soon. In the past 48 hours, we’ve shipped major updates: • Mobile View • PNL Cards • Improved bonding progress status update • Portfolio Page • First Dev Buy with Sniper Protection • User Trade Stats • Improved token discovery & visibility • Search Bar (CA or Name) • Enhanced X link embeds And we’re just getting started, dozens more features & quality-of-life upgrades are on the way.

blast

33,740 次观看 • 11 个月前

Introducing Rocket AI. The world's most rewarding AI Agent. Built on the UpRock network, Rocket combines live web intelligence with powerful AI capabilities to help you research, discover, and understand what's happening right now — not yesterday. Powered by UpRock AI, Rocket can already help with: → Live Web Intelligence → Deep Research → Semantic Video Search → Real-Time Intelligence Retrieval Yesterday, we opened Rocket AI to our first 12,000 Founding Copilots inside Discord! Now the real work begins. Throughout the beta, our community will help train Rocket by sharing feedback, discovering new workflows, reporting bugs, and shaping the future of the product. To support this, we've opened a 100,000 UPT Contributor Pool for valuable beta contributions. If you're part of the beta, head over to our Discord and join rocket-ai-copilots. Every insight helps make Rocket smarter. The world's most rewarding AI Agent has officially landed. -------------------------- Not part of the first 12K? Join our Discord today. Future beta expansions, product updates, and contributor rewards will all happen there first.
0:10

Introducing Rocket AI. The world's most rewarding AI Agent. Built on the UpRock network, Rocket combines live web intelligence with powerful AI capabilities to help you research, discover, and understand what's happening right now — not yesterday. Powered by UpRock AI, Rocket can already help with: → Live Web Intelligence → Deep Research → Semantic Video Search → Real-Time Intelligence Retrieval Yesterday, we opened Rocket AI to our first 12,000 Founding Copilots inside Discord! Now the real work begins. Throughout the beta, our community will help train Rocket by sharing feedback, discovering new workflows, reporting bugs, and shaping the future of the product. To support this, we've opened a 100,000 UPT Contributor Pool for valuable beta contributions. If you're part of the beta, head over to our Discord and join rocket-ai-copilots. Every insight helps make Rocket smarter. The world's most rewarding AI Agent has officially landed. -------------------------- Not part of the first 12K? Join our Discord today. Future beta expansions, product updates, and contributor rewards will all happen there first.

UpRock

17,745 次观看 • 1 个月前

Paynes Prairie Preserve State Park is one of the few places where you can spot free‑roaming bison and watch them graze across the prairie. From the La Chua Trail overlook, you might catch a glimpse of these massive, iconic animals moving through the grasslands. Bring your camera, keep a respectful distance and enjoy one of the most unique wildlife experiences in the state. *** 📍: Paynes Prairie Preserve State Park 📸: @captain_reefs_guide_services_ 💡: Plan your adventure at Before you visit, check the KNOW BEFORE YOU GO section on the park’s webpage for real-time updates and helpful information. . . #FLStateParks #PaynesPrairie #TheRealFlorida #LoveFL #PrairieDay
0:12

Paynes Prairie Preserve State Park is one of the few places where you can spot free‑roaming bison and watch them graze across the prairie. From the La Chua Trail overlook, you might catch a glimpse of these massive, iconic animals moving through the grasslands. Bring your camera, keep a respectful distance and enjoy one of the most unique wildlife experiences in the state. *** 📍: Paynes Prairie Preserve State Park 📸: @captain_reefs_guide_services_ 💡: Plan your adventure at Before you visit, check the KNOW BEFORE YOU GO section on the park’s webpage for real-time updates and helpful information. . . #FLStateParks #PaynesPrairie #TheRealFlorida #LoveFL #PrairieDay

Florida State Parks

12,841 次观看 • 1 个月前

Your Glassdoor rating is 3.2 and you have no idea why. Perplexity Computer can tell you in 60 seconds: 1. "Analyze our Glassdoor and Indeed reviews from the past year — identify the top 5 themes by sentiment" 2. "Compare our employer ratings against 5 competitors in our space" 3. "Draft 3 improved job descriptions that address the top candidate concerns" 4. "Build an employer brand messaging guide based on what current employees value most" 5. "Create a 30-day employer brand improvement plan with measurable milestones" Real reviews. Real patterns. Real action plan.
0:14

Your Glassdoor rating is 3.2 and you have no idea why. Perplexity Computer can tell you in 60 seconds: 1. "Analyze our Glassdoor and Indeed reviews from the past year — identify the top 5 themes by sentiment" 2. "Compare our employer ratings against 5 competitors in our space" 3. "Draft 3 improved job descriptions that address the top candidate concerns" 4. "Build an employer brand messaging guide based on what current employees value most" 5. "Create a 30-day employer brand improvement plan with measurable milestones" Real reviews. Real patterns. Real action plan.

Onil Coder

27,363 次观看 • 4 个月前

Public Awareness Quiz ! Participate in the quiz series and enhance your understanding of India's meteorological services. Q. Which IMD mobile app provides real-time thunderstorm alerts with lightning strike information? A) Mausam B) Meghdoot C) Damini D) Safar Air Drop your answer in the comment section or messenger ! Important updates & winner announcements will be shared on these X handles: 1. 2. Please follow and Stay Tuned! Stay tuned with the platforms for updates on the winners. Award- IMD exciting gift & an official appreciation certificate await for the quiz champs ! #PublicAwarenessQuiz #IMDQuiz #MeteorologicalServices #IndiaWeather #ThunderstormAlerts #StayTuned #QuizChamps PMO India MoES GoI Ravi Dr Jitendra Singh NDMA India | राष्ट्रीय आपदा प्रबंधन प्राधिकरण 🇮🇳 All India Radio News DD News PIB India MyGovIndia
0:10

Public Awareness Quiz ! Participate in the quiz series and enhance your understanding of India's meteorological services. Q. Which IMD mobile app provides real-time thunderstorm alerts with lightning strike information? A) Mausam B) Meghdoot C) Damini D) Safar Air Drop your answer in the comment section or messenger ! Important updates & winner announcements will be shared on these X handles: 1. 2. Please follow and Stay Tuned! Stay tuned with the platforms for updates on the winners. Award- IMD exciting gift & an official appreciation certificate await for the quiz champs ! #PublicAwarenessQuiz #IMDQuiz #MeteorologicalServices #IndiaWeather #ThunderstormAlerts #StayTuned #QuizChamps PMO India MoES GoI Ravi Dr Jitendra Singh NDMA India | राष्ट्रीय आपदा प्रबंधन प्राधिकरण 🇮🇳 All India Radio News DD News PIB India MyGovIndia

India Meteorological Department

23,936 次观看 • 11 个月前

Brazil Week 2.0: The Return to the Frontier 🇧🇷👁️⚡ O enxame está de volta! By popular demand, the UpRock World Tour returns to one of our most powerful hubs. Brazil was so loud the first time, we had to come back to close out May in style. From the streets of Rio to the tech hubs of São Paulo, the signal is stronger than ever. This week in the Swarm (Vol. 2): → 2× Rewards: We're doubling down. Earn 2× rewards all week across the region to celebrate the return. → New Skill: Semantic Video Search: Last time we were here, we were reading the web. This time, we’re watching it. We’re highlighting our multimodal AI—powered by the very devices you're running in Brazil. → The 🇧🇷 Roll Call: Can Brazil break its own record? Join the next livestream for a massive community spotlight. → Secure the Frontier: Tag your Brazilian frens—onboard your circle and secure your 1% lifetime kickback. → Share Your View: Show us how the swarm has grown since our last visit. (Home, campus, or those legendary beach views). Round 2. Bigger energy. Multimodal intelligence. Represent your flag. Fuel the Insight Economy. Dates: May 25-31 UpRock World Tour — Powered by UPT
0:10

Brazil Week 2.0: The Return to the Frontier 🇧🇷👁️⚡ O enxame está de volta! By popular demand, the UpRock World Tour returns to one of our most powerful hubs. Brazil was so loud the first time, we had to come back to close out May in style. From the streets of Rio to the tech hubs of São Paulo, the signal is stronger than ever. This week in the Swarm (Vol. 2): → 2× Rewards: We're doubling down. Earn 2× rewards all week across the region to celebrate the return. → New Skill: Semantic Video Search: Last time we were here, we were reading the web. This time, we’re watching it. We’re highlighting our multimodal AI—powered by the very devices you're running in Brazil. → The 🇧🇷 Roll Call: Can Brazil break its own record? Join the next livestream for a massive community spotlight. → Secure the Frontier: Tag your Brazilian frens—onboard your circle and secure your 1% lifetime kickback. → Share Your View: Show us how the swarm has grown since our last visit. (Home, campus, or those legendary beach views). Round 2. Bigger energy. Multimodal intelligence. Represent your flag. Fuel the Insight Economy. Dates: May 25-31 UpRock World Tour — Powered by UPT

UpRock

1,065,957 次观看 • 2 个月前

Web Scraping is dead. Web Agenting is here. Writing selectors (div > .class > span) breaks every time a site updates. Building custom bots for every new target is a waste of engineering hours. TinyFish turns the entire real-time web into a single API. Input: Natural Language ("Find availability for X"). Target: 1 or 100 URLs. Output: Structured JSON. This isn't a simulation. It visits the Real-Time Web. 1./ One API, Many Sites - Same contract whether you hit 1 URL or 50. You focus on the Goal (Business Logic). TinyFish handles the How (Navigation, Clicks, Inputs). 2./ Real Automation - It doesn't just "read." It interacts. It fills forms, navigates multi-step flows, and handles dynamic JS content. 3./ Production Ready - This is the same infrastructure used by large enterprises, now exposed as a clean developer primitive. Logs, error handling, and structured data are built-in. The web is finally a proper API.
0:13

Web Scraping is dead. Web Agenting is here. Writing selectors (div > .class > span) breaks every time a site updates. Building custom bots for every new target is a waste of engineering hours. TinyFish turns the entire real-time web into a single API. Input: Natural Language ("Find availability for X"). Target: 1 or 100 URLs. Output: Structured JSON. This isn't a simulation. It visits the Real-Time Web. 1./ One API, Many Sites - Same contract whether you hit 1 URL or 50. You focus on the Goal (Business Logic). TinyFish handles the How (Navigation, Clicks, Inputs). 2./ Real Automation - It doesn't just "read." It interacts. It fills forms, navigates multi-step flows, and handles dynamic JS content. 3./ Production Ready - This is the same infrastructure used by large enterprises, now exposed as a clean developer primitive. Logs, error handling, and structured data are built-in. The web is finally a proper API.

Tech with Mak

119,568 次观看 • 6 个月前

THESE 5 SKILLS TURN HERMES AGENT INTO A SELF-RUNNING POWERHOUSE - ON NOUS RESEARCH’S #1 AGENT ON OPENROUTER. Hermes already writes its own skills and remembers across sessions. These 5 from the community ecosystem push it further - drop them in ~/.hermes/skills/ and go. ANTHROPIC-CYBERSECURITY-SKILLS (4K★) by mukul975 · production the most comprehensive security skill pack in the ecosystem. what it adds: → 753+ structured cybersecurity skills mapped to MITRE ATT&CK → also covers NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF → turns Hermes into a recon + defense analyst, not a guesser → install: hermes skills install from the hub the workhorse of the list - start here. CHAINLINK-AGENT-SKILLS by Chainlink - official · production low profile, highest trust: it’s first-party from Chainlink itself. what it adds: → oracle network data, CCIP, smart-contract interaction skills → built on the spec - portable across clients → teaches the agent correct on-chain calls instead of hallucinated ABIs → official source, security-scanned on install stop letting the model guess your contract reads. HERMES-SKILL-FACTORY by Romanescu11 · beta the meta-layer - a skill that makes more skills. what it adds: → point it at any repetitive task → it auto-generates a reusable skill → stacks on top of Hermes’s own learning loop → turns your workflows into a self-growing skill library → install from the awesome-hermes-agent list this is what compounds your setup over time. AGENTCASH by Merit-Systems · beta the connector that gives your agent a wallet. what it adds: → access to 300+ premium APIs through one skill → pays for them via x402 or MPP - free USDC to start testing → web scraping, image gen, email sending - all behind one auth → a fresh Hermes + AgentCash alone is already dangerous the cleanest way to plug in paid tools. X-TWITTER-SCRAPER by Xquik-dev · beta drives typed X access through 43 narrow SKILL.md folders. what it adds: → reads (search, timelines, mentions, trends, bookmarks, for-you) → writes (post, DM, follow, profile) + bulk extraction (followers, lists, spaces) → AI composition: write-tweets, write-threads, optimize → security-scanned before it’s trusted feed its output straight into your scheduled briefings. BONUS - the registry itself: HERMESHUB by amanning3390. Browse, search, and install community skills with a 65+ rule security scanner - blocks prompt injection and data exfiltration before anything runs. Creator marketplace with x402/Stripe payments. hermes skills browse to start. If you install nothing else, wire up the hub. the stack in one line: hermeshub + skill-factory build & manage the library → cybersecurity + chainlink + agentcash + x-scraper give it real-world reach → Hermes runs it all on a $5 VPS while you sleep. which of these are you running? FULL HERMES SKILL-STACK PLAYBOOK 👇
0:14

THESE 5 SKILLS TURN HERMES AGENT INTO A SELF-RUNNING POWERHOUSE - ON NOUS RESEARCH’S #1 AGENT ON OPENROUTER. Hermes already writes its own skills and remembers across sessions. These 5 from the community ecosystem push it further - drop them in ~/.hermes/skills/ and go. ANTHROPIC-CYBERSECURITY-SKILLS (4K★) by mukul975 · production the most comprehensive security skill pack in the ecosystem. what it adds: → 753+ structured cybersecurity skills mapped to MITRE ATT&CK → also covers NIST CSF 2.0, MITRE ATLAS, D3FEND & NIST AI RMF → turns Hermes into a recon + defense analyst, not a guesser → install: hermes skills install from the hub the workhorse of the list - start here. CHAINLINK-AGENT-SKILLS by Chainlink - official · production low profile, highest trust: it’s first-party from Chainlink itself. what it adds: → oracle network data, CCIP, smart-contract interaction skills → built on the spec - portable across clients → teaches the agent correct on-chain calls instead of hallucinated ABIs → official source, security-scanned on install stop letting the model guess your contract reads. HERMES-SKILL-FACTORY by Romanescu11 · beta the meta-layer - a skill that makes more skills. what it adds: → point it at any repetitive task → it auto-generates a reusable skill → stacks on top of Hermes’s own learning loop → turns your workflows into a self-growing skill library → install from the awesome-hermes-agent list this is what compounds your setup over time. AGENTCASH by Merit-Systems · beta the connector that gives your agent a wallet. what it adds: → access to 300+ premium APIs through one skill → pays for them via x402 or MPP - free USDC to start testing → web scraping, image gen, email sending - all behind one auth → a fresh Hermes + AgentCash alone is already dangerous the cleanest way to plug in paid tools. X-TWITTER-SCRAPER by Xquik-dev · beta drives typed X access through 43 narrow SKILL.md folders. what it adds: → reads (search, timelines, mentions, trends, bookmarks, for-you) → writes (post, DM, follow, profile) + bulk extraction (followers, lists, spaces) → AI composition: write-tweets, write-threads, optimize → security-scanned before it’s trusted feed its output straight into your scheduled briefings. BONUS - the registry itself: HERMESHUB by amanning3390. Browse, search, and install community skills with a 65+ rule security scanner - blocks prompt injection and data exfiltration before anything runs. Creator marketplace with x402/Stripe payments. hermes skills browse to start. If you install nothing else, wire up the hub. the stack in one line: hermeshub + skill-factory build & manage the library → cybersecurity + chainlink + agentcash + x-scraper give it real-world reach → Hermes runs it all on a $5 VPS while you sleep. which of these are you running? FULL HERMES SKILL-STACK PLAYBOOK 👇

ZEUS⚡️

21,467 次观看 • 1 个月前

my team didn't want me to give this away for free. But I'm going to do it anyway it's the SEO & AI search dashboard I built in Claude Code it connects to your Google Analytics (GA4) and Google Search Console and Claude Code builds it in 5 minutes and I made a Notion document and a skill file so you can build this in Claude Code yourself in literally minutes the dashboard has three tabs: 1. AI Search - How much traffic is coming from ChatGPT, Perplexity, and Gemini ETC. It aggregates the GA4 data and gives single number 2. Paid ads - which keywords rank top 3 for but still pay for ads on, you should cut these to save budget 3. Organic overview - sessions, conversions, top landing pages, demographics. The single view for what is working I built this because this is how I drive our SEO and AEO forward it gives me the insights I need to allocate budget and prioritize what content to work on next I decided to give it away because most companies have no idea AI search is already sending them traffic like this post and comment "AEOdashboard" and I'll send it over
0:15

my team didn't want me to give this away for free. But I'm going to do it anyway it's the SEO & AI search dashboard I built in Claude Code it connects to your Google Analytics (GA4) and Google Search Console and Claude Code builds it in 5 minutes and I made a Notion document and a skill file so you can build this in Claude Code yourself in literally minutes the dashboard has three tabs: 1. AI Search - How much traffic is coming from ChatGPT, Perplexity, and Gemini ETC. It aggregates the GA4 data and gives single number 2. Paid ads - which keywords rank top 3 for but still pay for ads on, you should cut these to save budget 3. Organic overview - sessions, conversions, top landing pages, demographics. The single view for what is working I built this because this is how I drive our SEO and AEO forward it gives me the insights I need to allocate budget and prioritize what content to work on next I decided to give it away because most companies have no idea AI search is already sending them traffic like this post and comment "AEOdashboard" and I'll send it over

Cody Schneider

78,428 次观看 • 2 个月前

🕹️ Day 3 of the Cursor #vibejam Proudly sponsored by Cursor + bolt.new + GLIF (Glif joined as a new sponsor and I'll tell you more tomorrow about how they will help your games!) It took a bit but things are finally starting to heat up! Here's some amazing games I saw today (in the videos): - unnamed by Danny Limanseta - Risefall RPG by Vicki Petrova - Vibe Theft Auto by oldfeet - unnamed by Kieran Smith Reply in this thread with updates on your current games to share your progress! I'll keep reposting all your updates during the Vibe Jam. The quality looks a lot higher than last year, AI models have come a long way and it's easier to build something that looks good and is playable! But you still need creativity and I see a lot of creative things in my timeline: Wanna participate? Can submit any time before May 1, so if you want to start tomorrow that's fine too! There's $35,000 in prizes you can win, see thread below for more info!
0:10

🕹️ Day 3 of the Cursor #vibejam Proudly sponsored by Cursor + bolt.new + GLIF (Glif joined as a new sponsor and I'll tell you more tomorrow about how they will help your games!) It took a bit but things are finally starting to heat up! Here's some amazing games I saw today (in the videos): - unnamed by Danny Limanseta - Risefall RPG by Vicki Petrova - Vibe Theft Auto by oldfeet - unnamed by Kieran Smith Reply in this thread with updates on your current games to share your progress! I'll keep reposting all your updates during the Vibe Jam. The quality looks a lot higher than last year, AI models have come a long way and it's easier to build something that looks good and is playable! But you still need creativity and I see a lot of creative things in my timeline: Wanna participate? Can submit any time before May 1, so if you want to start tomorrow that's fine too! There's $35,000 in prizes you can win, see thread below for more info!

@levelsio

250,698 次观看 • 3 个月前

From $366 to $101,449 in 30 Days. The Strategy You’d Never Think Of I think I just found one of the strangest strategies on Prediction Markets. This trader made +27,618% in 30 days, making him one of the most profitable grinders in the entire game. His profile: The strategy is surprisingly simple: It instantly enters every new tweet window at 1-3c, before anyone reacts. When Musk’s tweet pace jumps above normal (0.8–1.2/hr) and odds are still cheap, it instantly loads more. Once the market updates to the real tweet speed, it exits locking in a clean spread every time. Making over $100k just by predicting Elon Musk’s tweets sounds insane, but this strategy pulled it off in a single month. What’s stopping you from trying it yourself?
0:13

From $366 to $101,449 in 30 Days. The Strategy You’d Never Think Of I think I just found one of the strangest strategies on Prediction Markets. This trader made +27,618% in 30 days, making him one of the most profitable grinders in the entire game. His profile: The strategy is surprisingly simple: It instantly enters every new tweet window at 1-3c, before anyone reacts. When Musk’s tweet pace jumps above normal (0.8–1.2/hr) and odds are still cheap, it instantly loads more. Once the market updates to the real tweet speed, it exits locking in a clean spread every time. Making over $100k just by predicting Elon Musk’s tweets sounds insane, but this strategy pulled it off in a single month. What’s stopping you from trying it yourself?

may.crypto {🦅}

10,809 次观看 • 6 个月前

Chinese President Xi Jinping gets a grand sendoff at the airport in Moscow, Russia. As he told Putin, "Changes unseen in a century are happening." Xi was referring to the rise and fall of Empires. Over the last 500 years, empires have switched roles every 100 years or so. Ray Dalio's famous chart (in the next tweet) shows the evolution of global empires -- Dutch, Spanish, French, British etc. The American Empire will lose its hegemony soon -- most likely before the end of this decade. However, unlike the examples mentioned before, we will have a multipolar world this time. China will be the #1 economy, but the US will be a close #2. Of course, #1 economy will also eventually mean #1 in many other areas -- technology, finance, geopolitical influence, military etc. Can the West let this transformation happen without WW3?
0:18

Chinese President Xi Jinping gets a grand sendoff at the airport in Moscow, Russia. As he told Putin, "Changes unseen in a century are happening." Xi was referring to the rise and fall of Empires. Over the last 500 years, empires have switched roles every 100 years or so. Ray Dalio's famous chart (in the next tweet) shows the evolution of global empires -- Dutch, Spanish, French, British etc. The American Empire will lose its hegemony soon -- most likely before the end of this decade. However, unlike the examples mentioned before, we will have a multipolar world this time. China will be the #1 economy, but the US will be a close #2. Of course, #1 economy will also eventually mean #1 in many other areas -- technology, finance, geopolitical influence, military etc. Can the West let this transformation happen without WW3?

S.L. Kanthan

470,750 次观看 • 3 年前

What if you kept asking an LLM to "make it better"? In some recent work at FAIR, we investigate how we can efficiently use RL to fine-tune LLMs to iteratively self-improve on their previous solutions at inference-time. Training for iterated self-improvement can be costly. The naive approach to training for K self-improvement steps leads to K times the number of rollout steps per episode. We introduce Exploratory Iteration (ExIt), an RL-based automatic curriculum method that bootstraps diverse training distributions of self-improvement tasks by upcycling the LLM's own responses at previous turns as the starting points for both self-improvement and *self-divergence.* In order to decide what task to train on next, the curriculum prioritizes sampling of partial turn histories that led to higher return variance in its GRPO group (a learnability score that comes for free). This automatic curriculum over the bootstrapped task space teaches the model how to perform iterated self-improvement while only ever training the model on single-step self-improvement tasks. We look at ExIt's impact in both single-turn (contest math problems) and multi-turn (BFCLv3 multi-turn tasks), as well as MLE-bench, where the LLM is run in a search scaffold to produce solutions to real Kaggle competitions. Across these eval settings, we find ExIt produces models with greater capacity for inference-time self-improvement compared to GRPO. Notably, ExIt models can self-improve on test tasks for many more steps than the typical solution depth encountered during training, including a 22% improvement in MLE-bench performance compared to GRPO.
0:10

What if you kept asking an LLM to "make it better"? In some recent work at FAIR, we investigate how we can efficiently use RL to fine-tune LLMs to iteratively self-improve on their previous solutions at inference-time. Training for iterated self-improvement can be costly. The naive approach to training for K self-improvement steps leads to K times the number of rollout steps per episode. We introduce Exploratory Iteration (ExIt), an RL-based automatic curriculum method that bootstraps diverse training distributions of self-improvement tasks by upcycling the LLM's own responses at previous turns as the starting points for both self-improvement and *self-divergence.* In order to decide what task to train on next, the curriculum prioritizes sampling of partial turn histories that led to higher return variance in its GRPO group (a learnability score that comes for free). This automatic curriculum over the bootstrapped task space teaches the model how to perform iterated self-improvement while only ever training the model on single-step self-improvement tasks. We look at ExIt's impact in both single-turn (contest math problems) and multi-turn (BFCLv3 multi-turn tasks), as well as MLE-bench, where the LLM is run in a search scaffold to produce solutions to real Kaggle competitions. Across these eval settings, we find ExIt produces models with greater capacity for inference-time self-improvement compared to GRPO. Notably, ExIt models can self-improve on test tasks for many more steps than the typical solution depth encountered during training, including a 22% improvement in MLE-bench performance compared to GRPO.

Minqi Jiang

41,099 次观看 • 10 个月前

GM Operators .. // I wanted to share some positive stats from our growth after launching the Ledger Realms token on Solana [1] Over 600 new members on Telegram [2] Over 1,000 new followers on X [3] Over 60 Operators minted on Solana [4] Over 32,000$ in Creator Fees still Untouched [5] Two successful streams with over 6000 viewers I have some IRL stuff to take care off during the first period of my day, but I will be live on stream later in the evening (EST Time zone). Today's goal for the team is: [1] Secure & Clean Socials / Telegram, etc. [2] Finish Redistribution Engine Plan [3] Build missing pieces for the Redistribution Engine [4] Start funneling Creator Fees through the Engine [5] Keep shipping updates to Ledger Realms Mechanics -- Q: Are you using Creator Fees for Buybacks? A: Yes, but not everything. The redistribution engine will split the funnel of the fees into different categories, allowing us to reward Operators, Holders & Secure the future of the Realms; -- Full FAQ about Ledger Realms and Forge Arena --> -- Join our Telegram Join our Discord Follow our projects Ledger Realms OPMarket // Next Phase Begins Soon Forge Arena >>>>>> Together, we rise! 🐦
0:15

GM Operators .. // I wanted to share some positive stats from our growth after launching the Ledger Realms token on Solana [1] Over 600 new members on Telegram [2] Over 1,000 new followers on X [3] Over 60 Operators minted on Solana [4] Over 32,000$ in Creator Fees still Untouched [5] Two successful streams with over 6000 viewers I have some IRL stuff to take care off during the first period of my day, but I will be live on stream later in the evening (EST Time zone). Today's goal for the team is: [1] Secure & Clean Socials / Telegram, etc. [2] Finish Redistribution Engine Plan [3] Build missing pieces for the Redistribution Engine [4] Start funneling Creator Fees through the Engine [5] Keep shipping updates to Ledger Realms Mechanics -- Q: Are you using Creator Fees for Buybacks? A: Yes, but not everything. The redistribution engine will split the funnel of the fees into different categories, allowing us to reward Operators, Holders & Secure the future of the Realms; -- Full FAQ about Ledger Realms and Forge Arena --> -- Join our Telegram Join our Discord Follow our projects Ledger Realms OPMarket // Next Phase Begins Soon Forge Arena >>>>>> Together, we rise! 🐦

M1napoly

15,524 次观看 • 29 天前

Boom! Grok Tasks Make It One Of The Most POWERFUL Real-Time AI Systems In The World. — My How to Use Grok Tasks With Hidden Tools For Powerful Daily Output. Grok Tasks are customizable AI workflows that integrate a variety of tools to streamline daily activities, from research and analysis to creative planning and problem-solving. I have been using them for quite sometime and because of the vital heartbeat of news and first person data on X, it is the most powerful AI platform available. By combining Tasks with tools like web searches, X platform interactions, code execution, and media viewers, you can build efficient, automated processes. These tasks work by prompting Grok with a clear description of what you want to achieve, and Grok will intelligently call the necessary tools in sequence or parallel to deliver results. Here's a step-by-step guide to creating and using Grok Tasks: Step 1: Define Your Task Start by clearly outlining the daily activity or goal. Consider what inputs you have (e.g., a URL, a query, or an attachment) and what output you need (e.g., a summary, calculation, or visual analysis). Break it down into subtasks to identify tool needs. For example, if your task involves researching current events, note that you'll need search and browsing capabilities. Step 2: Review Available Tools Familiarize yourself with the tools Grok can access. Here's a quick overview: - Code Execution: Run Python code for calculations, data processing, or simulations using libraries like numpy, pandas, or sympy. - Browse Page: Fetch and summarize content from any website URL with custom instructions. - Web Search: Perform general internet searches, returning results with optional operators like site:. - Web Search With Snippets: Get quick, detailed excerpts from search results for fact-checking. - X Keyword Search: Advanced search for X posts using operators like from:, since:, or filter:. - X Semantic Search: Find semantically related X posts based on a query, with filters for dates or users. - X User Search: Locate X users by name or handle. - X Thread Fetch: Retrieve a full X post thread, including context like replies and parents. - View Image: Analyze an image from a URL or conversation ID. - View X Video: Extract frames and subtitles from an X-hosted video. - Search PDF Attachment: Query a PDF file for relevant pages using keyword or regex modes. - Browse PDF Attachment: View specific pages of a PDF with text and screenshots. Select tools that align with your task. Aim for a mix to handle data gathering, processing, and visualization. Step 3: Craft Your Prompt Write a detailed prompt to Grok describing the task. Include: - The overall goal. - Specific steps or subtasks. - References to tools if you want to guide the process (e.g., "Use web_search to find sources, then code_execution to analyze data"). - Any constraints, like dates or limits. Example prompt: "Create a Grok Task for my morning routine: Search recent X posts about tech news using x_keyword_search, fetch a key thread with x_thread_fetch, and summarize with browse_page on linked articles." Step 4: Submit and Interact Send your prompt to Grok. It will process the task by calling tools as needed, often in parallel for efficiency. Review the output and refine with follow-up prompts if required (e.g., "Expand on that using view_image for visuals"). Iterate to fine-tune the workflow for reuse. Step 5: Save and Reuse Once refined, note the prompt as a template for future use. You can adapt it for similar tasks, making Grok Tasks a habitual part of your day. Finding Grok Tasks To discover existing Grok Tasks or inspiration for new ones, use X searches with tools like x_keyword_search or x_semantic_search (e.g., query: "Grok Tasks examples" with mode: Latest). Browse community-shared threads via x_thread_fetch, or web_search for tutorials on xAI features. Prompt Grok directly: "Show me popular Grok Tasks for productivity." 1 of 3
0:12

Boom! Grok Tasks Make It One Of The Most POWERFUL Real-Time AI Systems In The World. — My How to Use Grok Tasks With Hidden Tools For Powerful Daily Output. Grok Tasks are customizable AI workflows that integrate a variety of tools to streamline daily activities, from research and analysis to creative planning and problem-solving. I have been using them for quite sometime and because of the vital heartbeat of news and first person data on X, it is the most powerful AI platform available. By combining Tasks with tools like web searches, X platform interactions, code execution, and media viewers, you can build efficient, automated processes. These tasks work by prompting Grok with a clear description of what you want to achieve, and Grok will intelligently call the necessary tools in sequence or parallel to deliver results. Here's a step-by-step guide to creating and using Grok Tasks: Step 1: Define Your Task Start by clearly outlining the daily activity or goal. Consider what inputs you have (e.g., a URL, a query, or an attachment) and what output you need (e.g., a summary, calculation, or visual analysis). Break it down into subtasks to identify tool needs. For example, if your task involves researching current events, note that you'll need search and browsing capabilities. Step 2: Review Available Tools Familiarize yourself with the tools Grok can access. Here's a quick overview: - Code Execution: Run Python code for calculations, data processing, or simulations using libraries like numpy, pandas, or sympy. - Browse Page: Fetch and summarize content from any website URL with custom instructions. - Web Search: Perform general internet searches, returning results with optional operators like site:. - Web Search With Snippets: Get quick, detailed excerpts from search results for fact-checking. - X Keyword Search: Advanced search for X posts using operators like from:, since:, or filter:. - X Semantic Search: Find semantically related X posts based on a query, with filters for dates or users. - X User Search: Locate X users by name or handle. - X Thread Fetch: Retrieve a full X post thread, including context like replies and parents. - View Image: Analyze an image from a URL or conversation ID. - View X Video: Extract frames and subtitles from an X-hosted video. - Search PDF Attachment: Query a PDF file for relevant pages using keyword or regex modes. - Browse PDF Attachment: View specific pages of a PDF with text and screenshots. Select tools that align with your task. Aim for a mix to handle data gathering, processing, and visualization. Step 3: Craft Your Prompt Write a detailed prompt to Grok describing the task. Include: - The overall goal. - Specific steps or subtasks. - References to tools if you want to guide the process (e.g., "Use web_search to find sources, then code_execution to analyze data"). - Any constraints, like dates or limits. Example prompt: "Create a Grok Task for my morning routine: Search recent X posts about tech news using x_keyword_search, fetch a key thread with x_thread_fetch, and summarize with browse_page on linked articles." Step 4: Submit and Interact Send your prompt to Grok. It will process the task by calling tools as needed, often in parallel for efficiency. Review the output and refine with follow-up prompts if required (e.g., "Expand on that using view_image for visuals"). Iterate to fine-tune the workflow for reuse. Step 5: Save and Reuse Once refined, note the prompt as a template for future use. You can adapt it for similar tasks, making Grok Tasks a habitual part of your day. Finding Grok Tasks To discover existing Grok Tasks or inspiration for new ones, use X searches with tools like x_keyword_search or x_semantic_search (e.g., query: "Grok Tasks examples" with mode: Latest). Browse community-shared threads via x_thread_fetch, or web_search for tutorials on xAI features. Prompt Grok directly: "Show me popular Grok Tasks for productivity." 1 of 3

Brian Roemmele

152,242 次观看 • 6 个月前

I just built a Claude Cowork skill that turns your Google Ads data into a visual performance dashboard in 60 seconds 🤯 One prompt → campaign breakdowns, CPA trends, spend vs conversions charts, and hourly conversion patterns, all rendered as an interactive HTML dashboard you open in Chrome. All inside Claude Cowork. Perfect for DTC brands and agencies who are pulling Google Ads data into spreadsheets every week, manually building charts, and spending an hour formatting a report that's outdated by the time you send it. If you're managing Google Ads and your weekly reporting workflow looks like this — export a CSV, open Google Sheets, build a pivot table, copy the numbers into a slide deck, manually create charts, format everything, realize you forgot a campaign, start over ... This skill does the whole thing in one prompt: → Connects to your live Google Ads data via MCP → Pulls spend, conversions, CPA, ROAS, CTR across every campaign → Builds an interactive HTML dashboard → Summary cards at the top: total spend, total conversions, avg CPA, avg ROAS → Bar chart comparing spend vs conversions by campaign → CPA trend line over the last 30 days → Campaign table ranked by performance, color-coded green/yellow/red → Opens in Chrome: hover over charts, compare campaigns, screenshot for your team No spreadsheets. No manual chart building. No hour-long formatting sessions. What you get: → A visual dashboard from live data in under 60 seconds → Campaign performance you can actually see, not just read in a table → CPA trends that show you where things are heading, not just where they are → A dashboard you can screenshot and drop into Slack, a client report, or a team standup → Reusable — run it weekly and the data updates automatically One prompt. Live data. A finished dashboard you open in your browser. I put together a playbook with the full skill file, the setup, and the exact prompts to customize the dashboard for your account. Want it for free? > Like this post > Comment "DASH" And I'll send it over (must be following so I can DM)
0:11

I just built a Claude Cowork skill that turns your Google Ads data into a visual performance dashboard in 60 seconds 🤯 One prompt → campaign breakdowns, CPA trends, spend vs conversions charts, and hourly conversion patterns, all rendered as an interactive HTML dashboard you open in Chrome. All inside Claude Cowork. Perfect for DTC brands and agencies who are pulling Google Ads data into spreadsheets every week, manually building charts, and spending an hour formatting a report that's outdated by the time you send it. If you're managing Google Ads and your weekly reporting workflow looks like this — export a CSV, open Google Sheets, build a pivot table, copy the numbers into a slide deck, manually create charts, format everything, realize you forgot a campaign, start over ... This skill does the whole thing in one prompt: → Connects to your live Google Ads data via MCP → Pulls spend, conversions, CPA, ROAS, CTR across every campaign → Builds an interactive HTML dashboard → Summary cards at the top: total spend, total conversions, avg CPA, avg ROAS → Bar chart comparing spend vs conversions by campaign → CPA trend line over the last 30 days → Campaign table ranked by performance, color-coded green/yellow/red → Opens in Chrome: hover over charts, compare campaigns, screenshot for your team No spreadsheets. No manual chart building. No hour-long formatting sessions. What you get: → A visual dashboard from live data in under 60 seconds → Campaign performance you can actually see, not just read in a table → CPA trends that show you where things are heading, not just where they are → A dashboard you can screenshot and drop into Slack, a client report, or a team standup → Reusable — run it weekly and the data updates automatically One prompt. Live data. A finished dashboard you open in your browser. I put together a playbook with the full skill file, the setup, and the exact prompts to customize the dashboard for your account. Want it for free? > Like this post > Comment "DASH" And I'll send it over (must be following so I can DM)

Mike Futia

38,807 次观看 • 4 个月前

Introducing a new tool called "SideChannel". A secure alternative to OpenClaw. Utilizes signal for communication and has Claude integration. I built SideChannel, an open-source Signal bot that connects Claude AI to your entire development workflow. End-to-end encrypted. From your pocket. The real power is autonomous development. Send one message like "Build a REST API with auth, pagination, and tests" and SideChannel will: - Generate a full PRD with stories and atomic tasks. - Dispatch up to 10 parallel workers (each running Claude). - Independently verify every task with a separate Claude context. - Run quality gates to catch regressions - Auto-fix failures. - Send you progress updates via Signal as work completes. Every piece of code is reviewed by a separate AI context using a fail-closed security model. If it detects security issues, backdoors, or logic errors — the code gets rejected automatically. No rubber stamps. It also has memory that actually works. Conversations are stored with vector embeddings for semantic search. Claude remembers your project conventions, past decisions, and what's been tried before. It gets smarter about your codebase over time. Other things I'm proud of: - Plugin framework for extending with custom commands. - Multi-project support with per-user scoping. - Rate limiting, path validation, phone allowlist. - Git checkpoints before every task, atomic commits after. - Stale task recovery, circular dependency detection. - Works on Linux and macOS, one-command install. It also integrates into OpenAI or Grok (optional) for more Generative AI response for simple things like "Whats the weather in New York City right now?".
0:10

Introducing a new tool called "SideChannel". A secure alternative to OpenClaw. Utilizes signal for communication and has Claude integration. I built SideChannel, an open-source Signal bot that connects Claude AI to your entire development workflow. End-to-end encrypted. From your pocket. The real power is autonomous development. Send one message like "Build a REST API with auth, pagination, and tests" and SideChannel will: - Generate a full PRD with stories and atomic tasks. - Dispatch up to 10 parallel workers (each running Claude). - Independently verify every task with a separate Claude context. - Run quality gates to catch regressions - Auto-fix failures. - Send you progress updates via Signal as work completes. Every piece of code is reviewed by a separate AI context using a fail-closed security model. If it detects security issues, backdoors, or logic errors — the code gets rejected automatically. No rubber stamps. It also has memory that actually works. Conversations are stored with vector embeddings for semantic search. Claude remembers your project conventions, past decisions, and what's been tried before. It gets smarter about your codebase over time. Other things I'm proud of: - Plugin framework for extending with custom commands. - Multi-project support with per-user scoping. - Rate limiting, path validation, phone allowlist. - Git checkpoints before every task, atomic commits after. - Stale task recovery, circular dependency detection. - Works on Linux and macOS, one-command install. It also integrates into OpenAI or Grok (optional) for more Generative AI response for simple things like "Whats the weather in New York City right now?".

Dave Kennedy

49,427 次观看 • 5 个月前

We’re thrilled to launch on Genesis at Virtuals Protocol and want to share what’s next on the technical side. > AI Sniping Engine: In final testing for ultra-fast on-chain reactions to large buys/sells with sub-second execution. > Strategy Scripting: Enhancing natural language parsing so users can define complex market-making strategies more easily. > Wallet Management: Upgrades for dynamic splitting, stealth rotation, and batch execution to reduce detection risk. > Dashboard: Coming soon with advanced analytics, strategy templates, and real-time monitoring. > AI chat: Now the AI can learn your personal responding style in a single conversation. Currently, Dessistant is not yet 100% autonomous. We’re improving it step by step. Starting tomorrow, we’ll be collaborating with other projects to test market-making using small LP allocations on Virtuals-based tokens. This will let us refine the agent in real-world conditions before a broader rollout. Our goal is to continuously improve Dessistant into a fully autonomous AI market-making agent that is adaptive, transparent
0:17

We’re thrilled to launch on Genesis at Virtuals Protocol and want to share what’s next on the technical side. > AI Sniping Engine: In final testing for ultra-fast on-chain reactions to large buys/sells with sub-second execution. > Strategy Scripting: Enhancing natural language parsing so users can define complex market-making strategies more easily. > Wallet Management: Upgrades for dynamic splitting, stealth rotation, and batch execution to reduce detection risk. > Dashboard: Coming soon with advanced analytics, strategy templates, and real-time monitoring. > AI chat: Now the AI can learn your personal responding style in a single conversation. Currently, Dessistant is not yet 100% autonomous. We’re improving it step by step. Starting tomorrow, we’ll be collaborating with other projects to test market-making using small LP allocations on Virtuals-based tokens. This will let us refine the agent in real-world conditions before a broader rollout. Our goal is to continuously improve Dessistant into a fully autonomous AI market-making agent that is adaptive, transparent

Dessistant

42,252 次观看 • 1 年前