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

Must watch. Anthropic's released an hour long deep dive on prompt engineering. Covers: ▸ What makes a good prompt ▸ Handling ambiguity, reasoning paths, and edge cases ▸ Enterprise vs research prompting strategies ▸ Using personas, metaphors, and structured logic ▸ Jailbreaking, trust, and testing prompts at scale

Lior⚡

103,301 subscribers

66,290 просмотров • 1 год назад •via X (Twitter)

Образование Новости и политика Наука и технологии

Anya Rossi• Live Now

Private livecam show

Комментарии: 11

Фото профиля Lior⚡

Lior⚡1 год назад

Link: Featuring @alexalbert__ , @AmandaAskell, @zswitten

Фото профиля SecBriefs | Making Cybersecurity Simple

SecBriefs | Making Cybersecurity Simple1 год назад

🔍 From vulnerability assessment to exploitation, hacking follows a precise, predictable path. 🛤️ Knowledge is key to defense! 🛡️ Cybersecurity Dictionary for Everyone can help you understand every step of the process. Available on Amazon:

Фото профиля yuki@indiehacker

yuki@indiehacker1 год назад

Wow, that sounds super interesting—thanks for the heads-up! Prompt engineering is evolving so fast lately. I’m especially curious about the part on using personas and metaphors. Did you watch it already? Any standout takeaways?

Фото профиля Mori Mori Mori

Mori Mori Mori1 год назад

Is OpenAI or anthropic gonna have the better pod?

Фото профиля Dave Lalande

Dave Lalande1 год назад

As things unfold, I think prompt engineering will be the solution. For small task apps, whether for a Python app or HTML/JavaScript, the prompt and the AI will take care of it on the fly. Most people don't want to or won't need to code. They will prompt.

Фото профиля Arpit Adlakha

Arpit Adlakha1 год назад

This is helpful , need to go through this 🙏

Фото профиля Vishal

Vishal1 год назад

Definitely. Prompt engineering is quickly becoming a core skill.

Фото профиля r

r1 год назад

>openai pod

Фото профиля Tom Bielecki

Tom Bielecki1 год назад

Seven months ago

Фото профиля FortuneCookie123

FortuneCookie1231 год назад

This was done 7 months ago

Фото профиля Video.Irish

Video.Irish1 год назад

i found that you can just ask for a JSON

Похожие видео

I had a great time chatting with three of Anthropic's top prompt engineers: Amanda Askell, Zack Witten, and David Hershey. We covered a few topics like: - what makes a good prompt engineer - practical tips for writing better prompts - what's happening inside an LLM - jailbreaks - the future of prompt engineering - and more If you want to understand how we think about prompt engineering, this may be the best resource out there. Here's the entire one hour and sixteen minutes of our conversation. Enjoy!

I had a great time chatting with three of Anthropic's top prompt engineers: Amanda Askell, Zack Witten, and David Hershey. We covered a few topics like: - what makes a good prompt engineer - practical tips for writing better prompts - what's happening inside an LLM - jailbreaks - the future of prompt engineering - and more If you want to understand how we think about prompt engineering, this may be the best resource out there. Here's the entire one hour and sixteen minutes of our conversation. Enjoy!

Alex Albert

333,662 просмотров • 1 год назад

"A wooden house" by Claude Fable 5 (max) Max reasoning effort is impressive, this is using my safe system prompt that removes creativity but makes buildings more functional and it's impressive I'll keep testing it with other past system prompts, but so far so good

"A wooden house" by Claude Fable 5 (max) Max reasoning effort is impressive, this is using my safe system prompt that removes creativity but makes buildings more functional and it's impressive I'll keep testing it with other past system prompts, but so far so good

Angel 🌼

1,524,071 просмотров • 1 месяц назад

In this week's Stability Seconds, we’re getting you up to speed on how to use simple prompting techniques to create more precise images faster by not having to rewrite your prompt. Here’s how: 1️⃣ Use positive and negative prompts: Combine positive prompts to describe what should appear in the image, like “an editorial close-up,” and negative prompts to guide the model away from unwanted elements, such as “low resolution.” 2️⃣ Create prompt journals: When using Stable Image Ultra, try the prompt journal node to store a list of prompts directly into your workflow. This keeps your best prompts in one place, so you can easily reuse them or build on them later. You can find more prompting techniques and start creating with Stable Image Ultra here 👉

In this week's Stability Seconds, we’re getting you up to speed on how to use simple prompting techniques to create more precise images faster by not having to rewrite your prompt. Here’s how: 1️⃣ Use positive and negative prompts: Combine positive prompts to describe what should appear in the image, like “an editorial close-up,” and negative prompts to guide the model away from unwanted elements, such as “low resolution.” 2️⃣ Create prompt journals: When using Stable Image Ultra, try the prompt journal node to store a list of prompts directly into your workflow. This keeps your best prompts in one place, so you can easily reuse them or build on them later. You can find more prompting techniques and start creating with Stable Image Ultra here 👉

Stability AI

14,109 просмотров • 1 год назад

Most people think prompting is just writing better sentences Anthropic shows why that's wrong This 32-minute workshop breaks down how prompts are built for production Not prompt hacks Not "act as" Not another $300 course Actual systems Evals Edge cases Model migration Agent loops Good prompt is not one message It's a workflow the model can follow, test and improve This matters because every new claude opus or sonnet will still reward the same rules Watch the prompting playbook Bookmark it before your next bad ai answer is just a bad setup

Most people think prompting is just writing better sentences Anthropic shows why that's wrong This 32-minute workshop breaks down how prompts are built for production Not prompt hacks Not "act as" Not another $300 course Actual systems Evals Edge cases Model migration Agent loops Good prompt is not one message It's a workflow the model can follow, test and improve This matters because every new claude opus or sonnet will still reward the same rules Watch the prompting playbook Bookmark it before your next bad ai answer is just a bad setup

rari

390,471 просмотров • 1 месяц назад

🚨GROK PROMPT FRAMEWORKS ARE GOING VIRAL, AND FOR GOOD REASON A new visual breaking down Grok prompt frameworks is making the rounds, showing how structured prompts like R-T-F, T-A-G, B-A-B, and R-I-S-E can massively improve AI outputs. The idea is simple: better prompts = better results, and Grok is being positioned less like a chatbot and more like a serious productivity tool. Most people blame AI when answers are weak, but the real bottleneck is how poorly they ask questions… but frameworks can fix that. Prompting isn’t fluff anymore, it’s a skill, and Grok users who learn it early get an edge. Source: Elon Musk , Video made with Grok Imagine

🚨GROK PROMPT FRAMEWORKS ARE GOING VIRAL, AND FOR GOOD REASON A new visual breaking down Grok prompt frameworks is making the rounds, showing how structured prompts like R-T-F, T-A-G, B-A-B, and R-I-S-E can massively improve AI outputs. The idea is simple: better prompts = better results, and Grok is being positioned less like a chatbot and more like a serious productivity tool. Most people blame AI when answers are weak, but the real bottleneck is how poorly they ask questions… but frameworks can fix that. Prompting isn’t fluff anymore, it’s a skill, and Grok users who learn it early get an edge. Source: Elon Musk , Video made with Grok Imagine

Mario Nawfal

320,781 просмотров • 6 месяцев назад

This clip is a great example of what real-world mature autonomy looks like. Apollo Go encounters an accident scene at night on a narrow street — low light, limited space, unpredictable conditions — and responds with patience and caution instead of rushing. Moments like this show why real-world data matters. Flexible decision-making, situational awareness, and safe handling of edge cases are what make autonomous driving work at scale. Baidu Inc.

This clip is a great example of what real-world mature autonomy looks like. Apollo Go encounters an accident scene at night on a narrow street — low light, limited space, unpredictable conditions — and responds with patience and caution instead of rushing. Moments like this show why real-world data matters. Flexible decision-making, situational awareness, and safe handling of edge cases are what make autonomous driving work at scale. Baidu Inc.

Parul Gautam

21,607 просмотров • 6 месяцев назад

Nimit Sohoni ( Nimit Sohoni ) is an ex-Citadel quant and AI researcher at Cartesia with a PhD from Stanford. I interviewed him to break down what both those careers paths are like. We discussed: • How SWEs can transition into AI research or quant roles • Comparing and contrasting tech vs finance experiences • What his current AI research work at Cartesia is like • If a PhD is necessary for either of these career paths • How to learn more about AI research and find papers • Developing good research taste • Finance work culture, NDAs, what the work is like Where to watch: • YouTube: • Spotify: • Apple Podcasts: • Transcript:

Nimit Sohoni ( Nimit Sohoni ) is an ex-Citadel quant and AI researcher at Cartesia with a PhD from Stanford. I interviewed him to break down what both those careers paths are like. We discussed: • How SWEs can transition into AI research or quant roles • Comparing and contrasting tech vs finance experiences • What his current AI research work at Cartesia is like • If a PhD is necessary for either of these career paths • How to learn more about AI research and find papers • Developing good research taste • Finance work culture, NDAs, what the work is like Where to watch: • YouTube: • Spotify: • Apple Podcasts: • Transcript:

Ryan Peterman

125,135 просмотров • 5 месяцев назад

Our first short course with Anthropic! Building Towards Computer Use with Anthropic. This teaches you to build an LLM-based agent that uses a computer interface by generating mouse clicks and keystrokes. Computer Use is an important, emerging capability for LLMs that will let AI agents do many more tasks than were possible before, since it lets them interact with interfaces designed for humans to use, rather than only tools that provide explicit API access. I hope you will enjoy learning about it! This course is taught by Anthropic's Head of Curriculum, Colt_Steele. You'll learn to apply image reasoning and tool use to "use" a computer as follows: a model processes an image of the screen, analyzes it to understand what's going on, and navigates the computer via mouse clicks and keystrokes. This course goes through the key building blocks, and culminates in a demo of an AI assistant that uses a web browser to search for a research paper, downloads the PDF, and finally summarizes the paper for you. In detail, you’ll: - Learn about Anthropic's family of models, when to use which one, and make API requests to Claude - Use multi-modal prompts that combine text and image content blocks, and also work with streaming responses - Improve your prompting by using prompt templates, using XML to structure prompts, and providing examples - Implement prompt caching to reduce cost and latency - Apply tool-use to build a chatbot that can call different tools to respond to queries - See all these building blocks come together in Computer Use demo Please sign up here:

Our first short course with Anthropic! Building Towards Computer Use with Anthropic. This teaches you to build an LLM-based agent that uses a computer interface by generating mouse clicks and keystrokes. Computer Use is an important, emerging capability for LLMs that will let AI agents do many more tasks than were possible before, since it lets them interact with interfaces designed for humans to use, rather than only tools that provide explicit API access. I hope you will enjoy learning about it! This course is taught by Anthropic's Head of Curriculum, Colt_Steele. You'll learn to apply image reasoning and tool use to "use" a computer as follows: a model processes an image of the screen, analyzes it to understand what's going on, and navigates the computer via mouse clicks and keystrokes. This course goes through the key building blocks, and culminates in a demo of an AI assistant that uses a web browser to search for a research paper, downloads the PDF, and finally summarizes the paper for you. In detail, you’ll: - Learn about Anthropic's family of models, when to use which one, and make API requests to Claude - Use multi-modal prompts that combine text and image content blocks, and also work with streaming responses - Improve your prompting by using prompt templates, using XML to structure prompts, and providing examples - Implement prompt caching to reduce cost and latency - Apply tool-use to build a chatbot that can call different tools to respond to queries - See all these building blocks come together in Computer Use demo Please sign up here:

Andrew Ng

170,366 просмотров • 1 год назад

In Prompt Engineering for Vision Models, taught by Abby Jacques Verre and Caleb Kaiser of Comet , you’ll learn how to prompt and fine-tune vision models for personalized image generation, image editing, object detection and segmentation. The prompts you'll use for vision models could be text, point coordinates, or bounding boxes, depending on the model. You'll also learn to tune hyperparameters to shape the output. Models you'll use include Segment-Anything Model (SAM), OWL-ViT, and Stable Diffusion. You'll also learn to fine-tune Stable Diffusion to generate personalized images (say, an image of a specific person), using a handful of images for training. As an example of a multi-step workflow, you'll use OWL-ViT to detect an object based on a text prompt, then pass the bounding box to SAM to create a segmentation mask, and input that mask into Stable Diffusion to replace the original object with a new one based on a text prompt. Controlling vision models can be tricky; this course will teach prompting and fine-tuning techniques to get precise control over their output. Get started here:

In Prompt Engineering for Vision Models, taught by Abby Jacques Verre and Caleb Kaiser of Comet , you’ll learn how to prompt and fine-tune vision models for personalized image generation, image editing, object detection and segmentation. The prompts you'll use for vision models could be text, point coordinates, or bounding boxes, depending on the model. You'll also learn to tune hyperparameters to shape the output. Models you'll use include Segment-Anything Model (SAM), OWL-ViT, and Stable Diffusion. You'll also learn to fine-tune Stable Diffusion to generate personalized images (say, an image of a specific person), using a handful of images for training. As an example of a multi-step workflow, you'll use OWL-ViT to detect an object based on a text prompt, then pass the bounding box to SAM to create a segmentation mask, and input that mask into Stable Diffusion to replace the original object with a new one based on a text prompt. Controlling vision models can be tricky; this course will teach prompting and fine-tuning techniques to get precise control over their output. Get started here:

Andrew Ng

151,198 просмотров • 2 лет назад

v4.5 just dropped for Pro & Premier subscribers 🔥 A wider range of genres, richer vocals, & enhanced prompt understanding for songs that match your vision. What’s New: 🙌 Expanded genres & smarter mashups: More genre options — Blends like midwest emo + neosoul or EDM + folk come together seamlessly. 🎤 Enhanced voices: Vocals now hit harder — with more depth, emotion, and range. From intimate whispers to full-on power hooks, v4.5 delivers with feeling. 🔊 More complex, textured sound: v4.5 picks up the subtleties that make your music shine — layered instruments, tone shifts, and sonic details with depth. Prompts like “leaf textures” or “melodic whistling” now come through with clarity and dimension. 🧠 Better prompt adherence: Your words hit harder. Mood, vibe, instruments, and detail are captured with precision—so what you imagine is what you hear. ✍️ Prompt enhancement helper: Drop in a few tags or a rough idea, hit Enhance, and get a rich, fully-formed style prompt you can roll with or remix. 🎭 Upgraded Covers + Personas: Covers hold onto more melodic detail. Genre switching feels seamless. Personas better preserve the vibe and character of your track — and now… 🤝🏽 Covers + Personas can be combined: Remix voice, structure, and style all at once. It’s a whole new way to create. 📈 Extended song length: Previously 4 minutes, now create up to 8 minutes without using Extend. 🏆 Improved audio: Fuller, more balanced mixes with reduced shimmer and degradation — everything sounds better.

v4.5 just dropped for Pro & Premier subscribers 🔥 A wider range of genres, richer vocals, & enhanced prompt understanding for songs that match your vision. What’s New: 🙌 Expanded genres & smarter mashups: More genre options — Blends like midwest emo + neosoul or EDM + folk come together seamlessly. 🎤 Enhanced voices: Vocals now hit harder — with more depth, emotion, and range. From intimate whispers to full-on power hooks, v4.5 delivers with feeling. 🔊 More complex, textured sound: v4.5 picks up the subtleties that make your music shine — layered instruments, tone shifts, and sonic details with depth. Prompts like “leaf textures” or “melodic whistling” now come through with clarity and dimension. 🧠 Better prompt adherence: Your words hit harder. Mood, vibe, instruments, and detail are captured with precision—so what you imagine is what you hear. ✍️ Prompt enhancement helper: Drop in a few tags or a rough idea, hit Enhance, and get a rich, fully-formed style prompt you can roll with or remix. 🎭 Upgraded Covers + Personas: Covers hold onto more melodic detail. Genre switching feels seamless. Personas better preserve the vibe and character of your track — and now… 🤝🏽 Covers + Personas can be combined: Remix voice, structure, and style all at once. It’s a whole new way to create. 📈 Extended song length: Previously 4 minutes, now create up to 8 minutes without using Extend. 🏆 Improved audio: Fuller, more balanced mixes with reduced shimmer and degradation — everything sounds better.

Suno

439,658 просмотров • 1 год назад

This custom GPT writes Sora 2 prompts that actually work 🤯 Most people prompting Sora 2 are getting garbage output. The problem isn't Sora 2. It's your prompts. Here's what I figured out after generating 100+ UGC videos: Sora 2 needs hyper-specific instructions—shot composition, timing, camera movement, lighting, scene details. Most people prompt like this: "Make a UGC video of someone using my skincare product" And then wonder why the output looks like AI slop. What actually gets results: "9:16 vertical handheld shot, soft natural window lighting, female creator mid-20s in bathroom mirror holding [brand] serum at eye level, genuine testimonial energy, 0-3s hook showing application and subtle reaction, shallow depth of field, clean warm color grade" That's the gap between unusable content and scroll-stopping UGC. So I built a custom GPT that writes these prompts for you: → Tell it your product (supplement, skincare, SaaS, whatever) → It generates a detailed timeline-based Sora 2 prompt → Shot composition, camera angles, timing breakdowns—all included → Optimized for 9:16 TikTok/Reels format → Copy and paste directly into Sora 2 No guessing. No prompt engineering rabbit holes. No "why does this look terrible" frustration. Professional UGC prompts in 10 seconds. Want free access to the Sora 2 Prompt Generator GPT? > Comment "SORA" > Like this post And I'll send it over (must be following so I can DM)

Mike Futia

13,262 просмотров • 6 месяцев назад

This is next-level smart: An open-source platform that evaluates your prompts and automatically refines them based on the results. Of course, it feels obvious after you see it: • You write a prompt • The system evaluates it across different scenarios • Based on the results, it refines it to improve results I recorded a quick video to show you how it works. It's pretty cool stuff! Here are some of the problems and best practices for teams building AI applications: 1. Testing your prompts manually doesn't scale 2. Prompts should not be spread throughout the codebase 3. Non-technical people need easy access to your prompts 4. Prompts can always use a version history to track changes 5. Monitoring the performance of prompts overtime is critical Evaluating the prompts is what keeps me up at night from this list. Of all the conversations I've had with companies and people building AI applications, this is the area that's causing the most pain. Testing a prompt is difficult. Think about how you'd test the response of a model subjectively. What do you account for, "tone," "objectivity," "completeness," "creativity," "readability," etc.? Last week, I met the developers behind Latitude, an open-source prompt engineering platform trying to solve all of these issues. You can try the platform in two ways: • You can self-host the platform. Free and open-source. • If you want to try their online product, their free tier is huge. Here is the link: Thanks to the Latitude team for collaborating with me on this post, and congratulations on going live with their product!

This is next-level smart: An open-source platform that evaluates your prompts and automatically refines them based on the results. Of course, it feels obvious after you see it: • You write a prompt • The system evaluates it across different scenarios • Based on the results, it refines it to improve results I recorded a quick video to show you how it works. It's pretty cool stuff! Here are some of the problems and best practices for teams building AI applications: 1. Testing your prompts manually doesn't scale 2. Prompts should not be spread throughout the codebase 3. Non-technical people need easy access to your prompts 4. Prompts can always use a version history to track changes 5. Monitoring the performance of prompts overtime is critical Evaluating the prompts is what keeps me up at night from this list. Of all the conversations I've had with companies and people building AI applications, this is the area that's causing the most pain. Testing a prompt is difficult. Think about how you'd test the response of a model subjectively. What do you account for, "tone," "objectivity," "completeness," "creativity," "readability," etc.? Last week, I met the developers behind Latitude, an open-source prompt engineering platform trying to solve all of these issues. You can try the platform in two ways: • You can self-host the platform. Free and open-source. • If you want to try their online product, their free tier is huge. Here is the link: Thanks to the Latitude team for collaborating with me on this post, and congratulations on going live with their product!

Santiago

64,146 просмотров • 1 год назад

24 ChatGPT-V Prompts for an Elite Personal Brand Top Personal Brand = Top $$$ Simple. So I use 24 ChatGPT-Vision Prompts to improve this: [🔖Bookmark & copy the best prompts for YOU] 1. Personal Branding 2. Content Creation 3. Client Engagement 4. Market Analysis 5. Networking Strategies 6. Consulting Expertise 7. Continuous Learning 8. Client Retention Strategies Let's deep dive! 🌟 ------ 1. Personal Branding Profile Picture Assessment Prompt: "Review the profile picture for my consulting brand. Analyze its professionalism, approachability, and alignment with my niche. Offer suggestions for a more impactful first impression." Branding Collaterals Evaluation Prompt: "Examine my business cards, letterheads, and other branding materials. Assess consistency, visual appeal, and the clarity of information. Provide design optimization recommendations." Website Design Insights Prompt: "Analyze the visual layout and design elements of my consulting website. Highlight areas of user engagement, navigational ease, and potential design bottlenecks. Suggest improvements for a seamless user experience." ------ 2. Content Creation Blog Post Imagery Feedback Prompt: "Review the images used in my blog posts. Assess their relevance, quality, and alignment with the content theme. Offer insights for more engaging visuals." Infographic Evaluation Prompt: "Inspect the infographic detailing a consulting concept. Check for clarity of data representation, design aesthetics, and information flow. Recommend any enhancements for better comprehension." Video Thumbnail Analysis Prompt: "Analyze the thumbnail for my video content. Gauge its attention-grabbing capability, relevance to video content, and overall design. Suggest improvements for higher click-through rates." Slide Deck Design Insights Prompt: "Review the design of my presentation slides. Evaluate consistency, visual hierarchy, and clarity of information. Provide tips for a more engaging and effective presentation." ------ 3. Client Engagement Client Testimonial Presentation Prompt: "Evaluate the visual presentation of client testimonials on my website. Assess prominence, readability, and emotional impact. Offer design strategies to enhance credibility and trust." Proposal Layout Review Prompt: "Inspect the layout and design of my client proposals. Analyze structure, visual flow, and emphasis on key points. Recommend changes for a more persuasive proposal." Client Onboarding Material Insights Prompt: "Review the visual elements of my client onboarding materials. Gauge clarity, engagement, and comprehensiveness. Suggest enhancements for a smoother client integration process." ------ 4. Market Analysis Competitor Branding Analysis Prompt: "Analyze visual branding elements of key competitors in the consulting space. Identify strengths, differentiation points, and areas of improvement. Offer strategies for distinctive brand positioning." Trend Graph Interpretation Prompt: "Interpret the graph showcasing consulting industry trends. Decode growth areas, declining segments, and market saturation points. Provide insights for strategic positioning." Visual SWOT Analysis Prompt: "Examine the visual SWOT analysis of my consulting brand. Decode the highlighted strengths, weaknesses, opportunities, and threats. Suggest actionable strategies based on the analysis." ------ 5. Networking Strategies Event Banner Review Prompt: "Assess the design of my banner for networking events. Analyze branding prominence, message clarity, and visual appeal. Offer design tweaks for better audience engagement." Networking Pitch Slide Feedback Prompt: "Review the visual slides for my networking pitch. Gauge flow, key highlights, and overall presentation design. Suggest enhancements for a memorable pitch." Business Seminar Poster Insights Prompt: "Analyze the poster design for a seminar I'm hosting. Assess visual hierarchy, call-to-action prominence, and overall design aesthetics. Provide recommendations for higher attendee turnout." ------ 6. Consulting Expertise Demonstration Case Study Visualization Feedback Prompt: "Inspect the visual representation of a case study I conducted. Evaluate clarity of results, design consistency, and information flow. Offer insights for a more compelling case study presentation." Expertise Infographic Review Prompt: "Review the infographic detailing my consulting expertise and achievements. Gauge visual impact, data clarity, and design appeal. Suggest improvements for a stronger brand statement." Webinar Visual Assets Analysis Prompt: "Examine the visual assets used in my webinar. Assess alignment with content, visual engagement factors, and clarity. Provide recommendations for a more engaging webinar experience." ------ 7. Continuous Learning & Growth Certification Badge Display Prompt: "Review the display of my certification badges on my website and profiles. Offer suggestions for prominence, design integration, and credibility enhancement." Course Content Visualization Feedback Prompt: "Evaluate the visual content of the online course I've created. Assess design consistency, clarity, and engagement factor. Recommend enhancements for a better learner experience." Workshop Poster Insights Prompt: "Analyze the design of my workshop announcement poster. Gauge visual hierarchy, clarity of details, and overall aesthetics. Offer design tweaks for higher participant interest." ------ 8. Client Retention Strategies Client Feedback Form Review Prompt: "Inspect the design of my client feedback form. Assess ease of completion, clarity of questions, and design appeal. Suggest improvements for higher response rates." Visual Client Update Template Feedback Prompt: "Review the template I use for visual client updates. Evaluate design consistency, clarity of progress representation, and overall aesthetics -------------- That's it for today. Follow me Alex Northstar and come back tomorrow for more practical value solopreneur AI Insights & Education. Because you'll have more free time to spend building, earning and enjoying life, with AI. Share for a BONUS Prompt in the DMs & let me know your thoughts⬇️

Alex Northstar

19,496 просмотров • 2 лет назад

Today, we’re rolling out an updated Deep Think mode available in the Gemini app for Google AI Ultra subscribers. Here’s what you need to know: — Gemini 3 Deep Think mode pushes the boundaries of intelligence even further, delivering meaningful improvement in reasoning capabilities — It works by using advanced parallel thinking to explore multiple hypotheses simultaneously — This builds on the Gemini 2.5 Deep Think variants that recently achieved a gold-medal standard at this year's International Mathematical Olympiad and at the International Collegiate Programming Contest World Finals Ultra subscribers can try Deep Think today by selecting “Deep Think” in the prompt bar and Gemini 3 Pro in the model selector.

Today, we’re rolling out an updated Deep Think mode available in the Gemini app for Google AI Ultra subscribers. Here’s what you need to know: — Gemini 3 Deep Think mode pushes the boundaries of intelligence even further, delivering meaningful improvement in reasoning capabilities — It works by using advanced parallel thinking to explore multiple hypotheses simultaneously — This builds on the Gemini 2.5 Deep Think variants that recently achieved a gold-medal standard at this year's International Mathematical Olympiad and at the International Collegiate Programming Contest World Finals Ultra subscribers can try Deep Think today by selecting “Deep Think” in the prompt bar and Gemini 3 Pro in the model selector.

Google AI

169,108 просмотров • 7 месяцев назад

In this episode, OpenAI Chief Research Officer Mark Chen joins allen to flambé shrimp, cook Korean stew, and chat about being at the frontier of AI research: why scaling laws and pre-training still matter, how OpenAI chooses research bets and allocates compute, what it means to develop research taste, why evals are in crisis, how to avoid benchmark-maxing, and what it will take for models to handle long-horizon real-world work, multimodal reasoning, and eventually end-to-end AI research. Timestamps: 0:00 Intro 0:28 The Soup Story 1:52 From Trading to AI Research 3:21 How to Develop Research Taste 5:23 RL, Evals, and Superhuman Benchmarks 8:17 Cooking Begins on the Impulse Stove 8:53 Scaling Laws, Pre-Training, and Reasoning 12:33 OpenAI’s Research Roadmap and Compute Allocation 15:48 What Makes a Great Researcher 19:33 The Evals Crisis and Benchmark-Maxing 24:34 Jagged Intelligence, Context, and Long-Horizon Learning 27:14 Shrimp Flambé and New Research Bets 31:32 Multimodal Models and One Architecture 32:36 Vibe Researching and End-to-End AI Research 34:36 Failed Bets, Postmortems, and OpenAI’s Alpha 37:07 Final Taste Test 37:53 Overrated vs. Underrated AI Research 41:00 Closing

In this episode, OpenAI Chief Research Officer Mark Chen joins allen to flambé shrimp, cook Korean stew, and chat about being at the frontier of AI research: why scaling laws and pre-training still matter, how OpenAI chooses research bets and allocates compute, what it means to develop research taste, why evals are in crisis, how to avoid benchmark-maxing, and what it will take for models to handle long-horizon real-world work, multimodal reasoning, and eventually end-to-end AI research. Timestamps: 0:00 Intro 0:28 The Soup Story 1:52 From Trading to AI Research 3:21 How to Develop Research Taste 5:23 RL, Evals, and Superhuman Benchmarks 8:17 Cooking Begins on the Impulse Stove 8:53 Scaling Laws, Pre-Training, and Reasoning 12:33 OpenAI’s Research Roadmap and Compute Allocation 15:48 What Makes a Great Researcher 19:33 The Evals Crisis and Benchmark-Maxing 24:34 Jagged Intelligence, Context, and Long-Horizon Learning 27:14 Shrimp Flambé and New Research Bets 31:32 Multimodal Models and One Architecture 32:36 Vibe Researching and End-to-End AI Research 34:36 Failed Bets, Postmortems, and OpenAI’s Alpha 37:07 Final Taste Test 37:53 Overrated vs. Underrated AI Research 41:00 Closing

Latent.Space

62,533 просмотров • 21 дней назад

🚨 Breaking the Hype 🚨 Since #DeepSeek is all the rage, I decided to test what the fuss is about. And guess what? I “broke” it on my first try! 🤯 Here’s the simple prompt that sent DeepSeek into an endless logic loop, I had to manually cancel the reasoning loop 💬 If a train leaves New York at 2 PM traveling 60 mph and another train leaves Los Angeles at 3 PM traveling 70 mph, when do they meet? Looks like DeepSeek needs more 🚧 work on reasoning! 🚧 In comparison #ChatGPT (OpenAI) didn’t break a sweat. I didnt check the math but the reasoning is very simple and plausible. Anyone else trying to push its limits? Let’s share! #AI #PromptEngineering #TechFun

🚨 Breaking the Hype 🚨 Since #DeepSeek is all the rage, I decided to test what the fuss is about. And guess what? I “broke” it on my first try! 🤯 Here’s the simple prompt that sent DeepSeek into an endless logic loop, I had to manually cancel the reasoning loop 💬 If a train leaves New York at 2 PM traveling 60 mph and another train leaves Los Angeles at 3 PM traveling 70 mph, when do they meet? Looks like DeepSeek needs more 🚧 work on reasoning! 🚧 In comparison #ChatGPT (OpenAI) didn’t break a sweat. I didnt check the math but the reasoning is very simple and plausible. Anyone else trying to push its limits? Let’s share! #AI #PromptEngineering #TechFun

Dreams of Mars 🕊❤️🚀🌕

100,806 просмотров • 1 год назад

New short course: Prompt Engineering with Llama 2, built in collaboration with Meta AI at Meta, and taught by Amit Sangani! Meta's Llama 2 has been game-changing for AI. Building with open source lets you control your own data, scrutinize errors, update (or not) the models as you please, and work alongside the global community advancing open models. Llama isn't a single model, it's a collection of models. In this course, you'll: - Learn the differences between different Llama 2 flavors, and when to use each. - Prompt the Llama chat models -- you'll also see how Llama's instruction tags work -- so they can help you with day-to-day tasks, like writing or summarization. - Use advanced prompting, like few-shot prompting for classification, and chain-of-thought prompting for solving logic problems. - Use specialized models in the Llama collection for specific tasks, like Code Llama to help you write, analyze, and improve code, and Llama Guard, which checks prompts and model responses for harmful content. The course also touches on how to run Llama 2 locally on your own computer. I hope you’ll take this course and try out these powerful, open models!

New short course: Prompt Engineering with Llama 2, built in collaboration with Meta AI at Meta, and taught by Amit Sangani! Meta's Llama 2 has been game-changing for AI. Building with open source lets you control your own data, scrutinize errors, update (or not) the models as you please, and work alongside the global community advancing open models. Llama isn't a single model, it's a collection of models. In this course, you'll: - Learn the differences between different Llama 2 flavors, and when to use each. - Prompt the Llama chat models -- you'll also see how Llama's instruction tags work -- so they can help you with day-to-day tasks, like writing or summarization. - Use advanced prompting, like few-shot prompting for classification, and chain-of-thought prompting for solving logic problems. - Use specialized models in the Llama collection for specific tasks, like Code Llama to help you write, analyze, and improve code, and Llama Guard, which checks prompts and model responses for harmful content. The course also touches on how to run Llama 2 locally on your own computer. I hope you’ll take this course and try out these powerful, open models!

Andrew Ng

162,833 просмотров • 2 лет назад

I’m really proud of a lot of the shots I have been able to achieve lately with the adherence to prompts getting so good it’s a level of creative control I e been dying for and really starting to make me actual dreams come true for just myself working with these tools! Prompting is such a unique and creative aspect of what we all do in AI. Knowing when to prompt long form and when not to, what keyword tokens and how they are applied in sequence, structure and formatting, they all matter. Especially with the models not needing extensive prompts, that’s actually where you can shine with how you use it because sometimes simpler is better but in other instances extensive motion/movement detail can be the difference maker for a cool shot or a spectacular one. I’ve been focusing a lot on shot development prompts and structure testing and I’ve managed to find a sweet spot. I’m Gonna put together a little rundown on how I did some of mine next week for this film. 🙌😎🔥💯🎥

I’m really proud of a lot of the shots I have been able to achieve lately with the adherence to prompts getting so good it’s a level of creative control I e been dying for and really starting to make me actual dreams come true for just myself working with these tools! Prompting is such a unique and creative aspect of what we all do in AI. Knowing when to prompt long form and when not to, what keyword tokens and how they are applied in sequence, structure and formatting, they all matter. Especially with the models not needing extensive prompts, that’s actually where you can shine with how you use it because sometimes simpler is better but in other instances extensive motion/movement detail can be the difference maker for a cool shot or a spectacular one. I’ve been focusing a lot on shot development prompts and structure testing and I’ve managed to find a sweet spot. I’m Gonna put together a little rundown on how I did some of mine next week for this film. 🙌😎🔥💯🎥

Dustin Hollywood

15,964 просмотров • 1 год назад

At first, prompting seemed to be a temporary workaround for getting the most out of large language models. But over time, it's become critical to the way we interact with AI. On the Lightcone Podcast, Garry, Harj, Diana, and Jared break down what they've learned from working with hundreds of founders building with LLMs: why prompting still matters, where it breaks down, and how teams are making it more reliable in production. They share real examples of prompts that failed, how companies are testing for quality, and what the best teams are doing to make LLM outputs useful and predictable. 0:58 - Parahelp’s prompt example 4:59 - Different types of prompts 6:51 - Metaprompting 7:58 - Using examples 12:10 - Some tricks for longer prompts 14:18 - Findings on evals 17:25 - Every founder has become a forward-deployed engineer (FDE) 23:18 - Vertical AI agents are closing big deals with the FDE model 26:13 - The personalities of the different LLMs 27:26 - Lessons from rubrics 29:47 - Kaizen and the art of communication

At first, prompting seemed to be a temporary workaround for getting the most out of large language models. But over time, it's become critical to the way we interact with AI. On the Lightcone Podcast, Garry, Harj, Diana, and Jared break down what they've learned from working with hundreds of founders building with LLMs: why prompting still matters, where it breaks down, and how teams are making it more reliable in production. They share real examples of prompts that failed, how companies are testing for quality, and what the best teams are doing to make LLM outputs useful and predictable. 0:58 - Parahelp’s prompt example 4:59 - Different types of prompts 6:51 - Metaprompting 7:58 - Using examples 12:10 - Some tricks for longer prompts 14:18 - Findings on evals 17:25 - Every founder has become a forward-deployed engineer (FDE) 23:18 - Vertical AI agents are closing big deals with the FDE model 26:13 - The personalities of the different LLMs 27:26 - Lessons from rubrics 29:47 - Kaizen and the art of communication

Y Combinator

352,086 просмотров • 1 год назад