Loading video...

Video Failed to Load

Go Home

Grok modes. Lazy, speed, brains; everything in between. Pick one. Fast answers, step-by-step thinking, or multiple AIs teaming up on your problem. How to pick your mode: •⁠ ⁠Start a chat - app, or 𝕏. Nothing fancy. •⁠ ⁠Bottom bar - tap “Auto” next to “Ask anything”. That’s the...

MarioNawfal's profile picture

Mario Nawfal

3,848,052 subscribers

24,546 views • 5 months ago •via X (Twitter)

EducationScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Set up Grok to open with your iPhone’s Action Button... It’s insanely useful Whenever you need quick answers or help, just press and hold the Action Button Grok opens instantly in voice mode, so you can start talking right away, no app opening needed How to set it up: • Open Settings • Tap Action Button • Swipe to Shortcut • Select Talk to Grok (search for Grok) That’s it.... From now on, one press launches Grok in voice mode... Fast, hands-free, and always ready.....
0:35

Set up Grok to open with your iPhone’s Action Button... It’s insanely useful Whenever you need quick answers or help, just press and hold the Action Button Grok opens instantly in voice mode, so you can start talking right away, no app opening needed How to set it up: • Open Settings • Tap Action Button • Swipe to Shortcut • Select Talk to Grok (search for Grok) That’s it.... From now on, one press launches Grok in voice mode... Fast, hands-free, and always ready.....

X Freeze

20,813 views • 9 months ago

Just Grok it. Chat. Ask anything. Push it anywhere. Everyday questions, quick facts, or research. Stories or role-play. Brainstorming, problem-solving, dumb ideas that turn smart. All fair game. How to Grok it: •⁠ ⁠Download or update Grok first. Obviously. Open app, or 𝕏. •⁠ ⁠Enter your prompt; clear or chaotic •⁠ ⁠Add persona flavor; e.g., “be unhinged”, “act as a wizard”, or “Regular Mode” •⁠ ⁠Continue chatting. Grok remembers context in the same conversation.
0:21

Just Grok it. Chat. Ask anything. Push it anywhere. Everyday questions, quick facts, or research. Stories or role-play. Brainstorming, problem-solving, dumb ideas that turn smart. All fair game. How to Grok it: •⁠ ⁠Download or update Grok first. Obviously. Open app, or 𝕏. •⁠ ⁠Enter your prompt; clear or chaotic •⁠ ⁠Add persona flavor; e.g., “be unhinged”, “act as a wizard”, or “Regular Mode” •⁠ ⁠Continue chatting. Grok remembers context in the same conversation.

Mario Nawfal

21,692 views • 5 months ago

I tested Gemini 3.1 Flash Live and one thing stood out immediately: the speed. You ask a question. It answers almost instantly. No awkward delay. No clunky back-and-forth. Then it gets more interesting. You can: use voice mode, use webcam mode, share your screen, connect tools, add system instructions, and publish the app to the cloud. So this is not just “voice chat with AI.” It is the beginning of real-time AI agents you can actually ship.
18:11

I tested Gemini 3.1 Flash Live and one thing stood out immediately: the speed. You ask a question. It answers almost instantly. No awkward delay. No clunky back-and-forth. Then it gets more interesting. You can: use voice mode, use webcam mode, share your screen, connect tools, add system instructions, and publish the app to the cloud. So this is not just “voice chat with AI.” It is the beginning of real-time AI agents you can actually ship.

Julian Goldie SEO

21,000 views • 6 months ago

Grok Build CLI just dropped. Not for dabbling. This is dev mode. Runs in your terminal and actually gets work done. Activate your personal dev mode. SuperGrok Heavy users only. Try it: * Confirm access - this is SuperGrok Heavy only, not your regular plan * Upgrade if needed - or * Install it - run the one-liner in your terminal * Jump into your project - cd into your repo * Launch Grok - type “grok” and log in once * Start prompting - audits, features, fixes, whatever you need * Use plan mode - review tasks and diffs before anything runs * Spin up sub-agents - let it handle big workloads * Go headless - automate scripts when you don’t want to babysit * Keep iterating - refine, rerun, improve You’re up and running in minutes. No excuses. Tip: Start with a small task in an existing repo to get comfortable with the plan and review workflow. xAI engineers are actively gathering feedback. Share your experience on 𝕏. xAI
0:33

Grok Build CLI just dropped. Not for dabbling. This is dev mode. Runs in your terminal and actually gets work done. Activate your personal dev mode. SuperGrok Heavy users only. Try it: * Confirm access - this is SuperGrok Heavy only, not your regular plan * Upgrade if needed - or * Install it - run the one-liner in your terminal * Jump into your project - cd into your repo * Launch Grok - type “grok” and log in once * Start prompting - audits, features, fixes, whatever you need * Use plan mode - review tasks and diffs before anything runs * Spin up sub-agents - let it handle big workloads * Go headless - automate scripts when you don’t want to babysit * Keep iterating - refine, rerun, improve You’re up and running in minutes. No excuses. Tip: Start with a small task in an existing repo to get comfortable with the plan and review workflow. xAI engineers are actively gathering feedback. Share your experience on 𝕏. xAI

Mario Nawfal

75,474 views • 4 months ago

Here’s how AI Mode and Google Maps’ Ask Maps can help you find and book the right table at the right restaurant right when you want it. Here’s how: 1️⃣ Describe all your needs — including details about your group size, ideal vibe and time and dietary preferences — in AI Mode in Search or Google Maps’ Ask Maps. 2️⃣ We’ll search across multiple reservation platforms and websites to find real-time availability for options that match your criteria. 3️⃣ From there, Ask Maps and AI Mode will present a curated list of restaurants to choose from with direct links to finalize your bookings through partners like OpenTable or Resy.
0:25

Here’s how AI Mode and Google Maps’ Ask Maps can help you find and book the right table at the right restaurant right when you want it. Here’s how: 1️⃣ Describe all your needs — including details about your group size, ideal vibe and time and dietary preferences — in AI Mode in Search or Google Maps’ Ask Maps. 2️⃣ We’ll search across multiple reservation platforms and websites to find real-time availability for options that match your criteria. 3️⃣ From there, Ask Maps and AI Mode will present a curated list of restaurants to choose from with direct links to finalize your bookings through partners like OpenTable or Resy.

Google

114,521 views • 5 months ago

BREAKING: OpenAI has officially released Agent Mode, allowing users to spin up a virtual desktop and take action. I got early access to ChatGPT Agent Mode and want to share my thoughts. Agent Mode lets you delegate tasks like searching or browsing to an agent inside a virtual desktop inside ChatGPT. You can spin up multiple little “workers” to fire off tasks and keep tabs as they chug through your to-do list AT THE SAME TIME. If it gets stuck or goes off track, you can take over manually. The UI is friendly and keeps the reasoning transparent, so you’re not left guessing what it’s doing or why. It’s not perfect and doesn’t give you as many artifacts as other players in the space, but a strong platform for “let’s see how much I can hand off and supervise”, and a first step toward managing fleets of AI workers instead of pinging a single chatbot for answers. If you’re on ChatGPT Pro, Plus, or Team, look for Agent Mode in the tools dropdown.
3:43

BREAKING: OpenAI has officially released Agent Mode, allowing users to spin up a virtual desktop and take action. I got early access to ChatGPT Agent Mode and want to share my thoughts. Agent Mode lets you delegate tasks like searching or browsing to an agent inside a virtual desktop inside ChatGPT. You can spin up multiple little “workers” to fire off tasks and keep tabs as they chug through your to-do list AT THE SAME TIME. If it gets stuck or goes off track, you can take over manually. The UI is friendly and keeps the reasoning transparent, so you’re not left guessing what it’s doing or why. It’s not perfect and doesn’t give you as many artifacts as other players in the space, but a strong platform for “let’s see how much I can hand off and supervise”, and a first step toward managing fleets of AI workers instead of pinging a single chatbot for answers. If you’re on ChatGPT Pro, Plus, or Team, look for Agent Mode in the tools dropdown.

Allie K. Miller

155,943 views • 1 year ago

5️⃣ Talk it out with Gemini Live Today, we’re bringing Deep Research into Gemini Live so you can launch comprehensive, multi-step research reports with your voice 🗣️✨ Using Deep Research, you can ask Gemini Live to research any topic, then step away, lock your screen, or switch tasks. Gemini works in the background and sends a notification when your report is ready. You can chat through the results, ask follow-up questions, and more. Learn more and sign up for your free student plan for one year →
0:51

5️⃣ Talk it out with Gemini Live Today, we’re bringing Deep Research into Gemini Live so you can launch comprehensive, multi-step research reports with your voice 🗣️✨ Using Deep Research, you can ask Gemini Live to research any topic, then step away, lock your screen, or switch tasks. Gemini works in the background and sends a notification when your report is ready. You can chat through the results, ask follow-up questions, and more. Learn more and sign up for your free student plan for one year →

Google

25,732 views • 1 month ago

Athena ≠ research mode. It's an entire execution environment for ASI:One. That means your personal AI gets somewhere to actually work. Instead of stopping at a summary you can read on the go, your personal AI can keep going until the in-depth documentation, presentation, or report is ready. Same intelligence. Deeper execution. Here’s how it works 👇
1:25

Athena ≠ research mode. It's an entire execution environment for ASI:One. That means your personal AI gets somewhere to actually work. Instead of stopping at a summary you can read on the go, your personal AI can keep going until the in-depth documentation, presentation, or report is ready. Same intelligence. Deeper execution. Here’s how it works 👇

Fetch.ai

18,273 views • 1 month ago

A SpaceXAI engineer just released a free 30-min course on Grok Bot Agents: from your first persistent agent to multi-agent chains that research, ship software, and hand work off to Cursor 05:26 - what actually lives inside a Grok agent 08:49 - the 3 ways to run agents 12:04 - chaining multiple agents into one workflow 16:06 - turning X followers into structured research inside Notion 23:08 - handing coding work from Grok straight to Cursor Most people still treat Grok like a chat window. It answers, you copy, you move on. The shift happens when the agents stop waiting for you: one finds the information, one structures it in Notion, one turns it into working code, and they pass the work down the line on their own. Thirty minutes here beats assembling your setup one agent at a time.
28:21

A SpaceXAI engineer just released a free 30-min course on Grok Bot Agents: from your first persistent agent to multi-agent chains that research, ship software, and hand work off to Cursor 05:26 - what actually lives inside a Grok agent 08:49 - the 3 ways to run agents 12:04 - chaining multiple agents into one workflow 16:06 - turning X followers into structured research inside Notion 23:08 - handing coding work from Grok straight to Cursor Most people still treat Grok like a chat window. It answers, you copy, you move on. The shift happens when the agents stop waiting for you: one finds the information, one structures it in Notion, one turns it into working code, and they pass the work down the line on their own. Thirty minutes here beats assembling your setup one agent at a time.

Morlex

16,710 views • 1 month ago

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 (acq. by Anaconda) for partnering with me on this post.
6:17

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 (acq. by Anaconda) for partnering with me on this post.

Santiago

73,304 views • 11 months ago

this is f**king insane. the open-source 100% FREE Perplexity replacement that runs entirely on your own hardware. the GitHub repo is at 36.7K stars, MIT license it lets you: → search the web, discussions, or academic papers → three search modes, including Quality mode for deep research → local models through Ollama, or OpenAI / Claude / Grok if you want → upload PDFs and ask questions about them → restrict a search to specific domains set GPT-6 Astra as the provider, switch to Quality mode, and you have a deep research loop running on your own box. that combination is not on sale anywhere. Perplexity Pro is $20 a month, this is one docker command repo below ↓
0:31

this is f**king insane. the open-source 100% FREE Perplexity replacement that runs entirely on your own hardware. the GitHub repo is at 36.7K stars, MIT license it lets you: → search the web, discussions, or academic papers → three search modes, including Quality mode for deep research → local models through Ollama, or OpenAI / Claude / Grok if you want → upload PDFs and ask questions about them → restrict a search to specific domains set GPT-6 Astra as the provider, switch to Quality mode, and you have a deep research loop running on your own box. that combination is not on sale anywhere. Perplexity Pro is $20 a month, this is one docker command repo below ↓

darkzodchi

41,733 views • 29 days ago

Got the vibe but not the words? Starting today, you can simply show or tell AI Mode about what you're looking for and get rich, visual results to explore. ✨ This experience works by bringing together Search’s world-class visual understanding from Lens and Image search with Gemini 2.5’s advanced multimodal capabilities. When you show AI Mode what you’re searching for it uses our new “visual search fan-out” technique to get a deeper understanding of what’s in an image (including subtle details and secondary objects), and runs multiple queries in the background. This helps it understand the visual context and nuance of your request to deliver relevant visual responses. You can also tell AI Mode what vibe you’re going for (no image needed). For example, ask AI Mode for bedroom design inspo and go back and forth to refine your vision. We’re rolling out this experience in English in the U.S. starting this week. Try it out and let us know what you think!
0:30

Got the vibe but not the words? Starting today, you can simply show or tell AI Mode about what you're looking for and get rich, visual results to explore. ✨ This experience works by bringing together Search’s world-class visual understanding from Lens and Image search with Gemini 2.5’s advanced multimodal capabilities. When you show AI Mode what you’re searching for it uses our new “visual search fan-out” technique to get a deeper understanding of what’s in an image (including subtle details and secondary objects), and runs multiple queries in the background. This helps it understand the visual context and nuance of your request to deliver relevant visual responses. You can also tell AI Mode what vibe you’re going for (no image needed). For example, ask AI Mode for bedroom design inspo and go back and forth to refine your vision. We’re rolling out this experience in English in the U.S. starting this week. Try it out and let us know what you think!

Google AI

68,285 views • 1 year ago

SHE CHARGES $7500 FOR A SITE YOU CONTROL BY WAVING AT YOUR WEBCAM no mouse, no scroll, no clicking anything, the camera tracks your hand and the whole site responds as you move it through the air how she built it: render the base clip in seedance 2 open replit and attach your design reference tell it to wire hand tracking with mediapipe map the hand position to the video in a single prompt still in the same chat, switch it to slide mode or a scroll transform whenever you want she never wrote one line of the tracking code, replit stood up the entire gesture layer from that one instruction meanwhile your local dev still thinks interactive means a button that lights up when the cursor grazes it the exact mediapipe prompt is in the guide below, save it before this becomes the standard
0:33

SHE CHARGES $7500 FOR A SITE YOU CONTROL BY WAVING AT YOUR WEBCAM no mouse, no scroll, no clicking anything, the camera tracks your hand and the whole site responds as you move it through the air how she built it: render the base clip in seedance 2 open replit and attach your design reference tell it to wire hand tracking with mediapipe map the hand position to the video in a single prompt still in the same chat, switch it to slide mode or a scroll transform whenever you want she never wrote one line of the tracking code, replit stood up the entire gesture layer from that one instruction meanwhile your local dev still thinks interactive means a button that lights up when the cursor grazes it the exact mediapipe prompt is in the guide below, save it before this becomes the standard

Bounce

59,981 views • 3 months ago

15 years of my X feed. Every interview I've ever done. Everything I've written on Real Vision. All of it now lives in one AI... and it talks in my voice. You can call it and ask what I'm thinking about this, or how you should think about that, or what this Everything Code actually is and how it works. It answers the way I would, pulling from everything Ive ever said. Intelligence used to be scarce... something you booked a meeting for. Now you can pick up the phone and talk to it whenever you like. That's The Exponential Age arriving in the most personal way possible. So go to and talk to me. Its free, its fun, and its me.
0:48

15 years of my X feed. Every interview I've ever done. Everything I've written on Real Vision. All of it now lives in one AI... and it talks in my voice. You can call it and ask what I'm thinking about this, or how you should think about that, or what this Everything Code actually is and how it works. It answers the way I would, pulling from everything Ive ever said. Intelligence used to be scarce... something you booked a meeting for. Now you can pick up the phone and talk to it whenever you like. That's The Exponential Age arriving in the most personal way possible. So go to and talk to me. Its free, its fun, and its me.

Raoul Pal

84,255 views • 7 days ago

Grok Bot + Kimi K3 can be turned into something bigger than an agent: an AI operating system the formula: AI OS = Router + Reasoning + Memory + Tools + Loops + Verification not one giant assistant. six layers that keep work moving without you step 1 -> Grok Bot becomes the operator. you give it the goal, it breaks the goal into jobs, assigns priorities and decides what part of the system should act next. step 2 -> Kimi K3 becomes the reasoning core. hard research, synthesis, long context and planning move here instead of forcing every task through the same model. step 3 -> externalize memory. store goals, decisions, failed attempts, artifacts and current state outside the chat. close the session, come back tomorrow, and the system still knows where it is. step 4 -> connect tools: search, code, files, APIs, docs and data. reasoning decides what should happen. tools actually make it happen. step 5 -> add the loop engine: plan -> execute -> inspect -> update memory -> retry. the loop can wait for new information, rerun a failed task, hand work to another agent or stop when the goal is complete. step 6 -> verify before output. tests, source checks, constraints and explicit completion rules decide whether the system ships the result or sends it back into the loop. that's the difference between an AI assistant and an AI operating system. an assistant waits for your next message. an operating system carries state, routes work and keeps moving. Grok Bot handles orchestration, Kimi K3 handles deeper reasoning, memory keeps the state alive, tools execute, the loop keeps the system running, verification decides when it is actually done. build one reliable loop and you have an agent. connect reasoning, memory, tools and multiple loops around it and you start building infrastructure. the full Grok Bot + Kimi K3 AI OS breakdown is below ↓
0:20

Grok Bot + Kimi K3 can be turned into something bigger than an agent: an AI operating system the formula: AI OS = Router + Reasoning + Memory + Tools + Loops + Verification not one giant assistant. six layers that keep work moving without you step 1 -> Grok Bot becomes the operator. you give it the goal, it breaks the goal into jobs, assigns priorities and decides what part of the system should act next. step 2 -> Kimi K3 becomes the reasoning core. hard research, synthesis, long context and planning move here instead of forcing every task through the same model. step 3 -> externalize memory. store goals, decisions, failed attempts, artifacts and current state outside the chat. close the session, come back tomorrow, and the system still knows where it is. step 4 -> connect tools: search, code, files, APIs, docs and data. reasoning decides what should happen. tools actually make it happen. step 5 -> add the loop engine: plan -> execute -> inspect -> update memory -> retry. the loop can wait for new information, rerun a failed task, hand work to another agent or stop when the goal is complete. step 6 -> verify before output. tests, source checks, constraints and explicit completion rules decide whether the system ships the result or sends it back into the loop. that's the difference between an AI assistant and an AI operating system. an assistant waits for your next message. an operating system carries state, routes work and keeps moving. Grok Bot handles orchestration, Kimi K3 handles deeper reasoning, memory keeps the state alive, tools execute, the loop keeps the system running, verification decides when it is actually done. build one reliable loop and you have an agent. connect reasoning, memory, tools and multiple loops around it and you start building infrastructure. the full Grok Bot + Kimi K3 AI OS breakdown is below ↓

Alex

13,312 views • 1 month ago

I built a knowledge/chat agent that actually sees inside your Shopify app. It knows every screen, every button, every text block, and every link — and can deep-link users directly into the right page of your live app. It’s called Zero Docs. ▻ Drop the agent directly into your live app ▻ Use our Chrome extension to capture your authenticated app screens ▻ It answers with real layout intelligence + deep links ▻ Reduce your support requests by 80% ▻ One click and it writes your docs too Deploy it in 10 minutes. Free to try. Plans start at $10/mo. First 50 Shopify apps get 4 free months of Bootstrap or Starter plans — just DM me or use code SHOPIFY4 at checkout. I'd love feedback from other Shopify devs — especially if you’ve tried other AI "chat" agents before.
8:14

I built a knowledge/chat agent that actually sees inside your Shopify app. It knows every screen, every button, every text block, and every link — and can deep-link users directly into the right page of your live app. It’s called Zero Docs. ▻ Drop the agent directly into your live app ▻ Use our Chrome extension to capture your authenticated app screens ▻ It answers with real layout intelligence + deep links ▻ Reduce your support requests by 80% ▻ One click and it writes your docs too Deploy it in 10 minutes. Free to try. Plans start at $10/mo. First 50 Shopify apps get 4 free months of Bootstrap or Starter plans — just DM me or use code SHOPIFY4 at checkout. I'd love feedback from other Shopify devs — especially if you’ve tried other AI "chat" agents before.

William Belk

1,078,580 views • 4 months ago

Jev + Opus 5.5: Anthropic's new model beats GPT-6 Astra for 1/5 the cost, and 4 API changes will 400 your agent before it writes a single line I pulled these 10 steps from the migration docs so you don't learn them in production step 1 → $4 / $20 per 1M. Opus 5 was $5 / $25. cache reads dropped from $0.50 to $0.20 step 2 → 66.4% on Terminal-Bench 4.0 vs GPT-6 Astra 57.9% and Opus 5 52.3%. +14.1 points in one release, and on FrontierCode it beats Astra at default effort for 1/5 the cost step 3 → thinking can't be turned off anymore. send thinking: disabled and you get a 400. drop the field, set effort step 4 → tool_choice any and tool are gone. 400. switch to auto + strict step 5 → edit anything above a thinking block and the request dies. append only, or opt into drop_block step 6 → computer_20251124 is dead on the API. 400. move to computer_toolset_20260801 step 7 → the quiet one: default effort fell from high to medium. your agent thinks less than you set it up to and nothing tells you step 8 → hop Opus 5.5 → Sonnet 5 → Opus 5.5 and you pay 4.36 instead of 3.32. +31%, the cache dies and Sonnet can't read Opus's reasoning step 9 → change effort at the top of the request and the cache is gone. Jev sets it per message and the cache stays step 10 → switch fast - standard mid-session and it's a full cache miss. Jev picks speed once, on turn one one model, three knobs, zero 400s. that is Jev + Opus 5.5 send this to your Claude Code before you touch the model ID, then read my full Jev deep dive in the article below ↓
0:40

Jev + Opus 5.5: Anthropic's new model beats GPT-6 Astra for 1/5 the cost, and 4 API changes will 400 your agent before it writes a single line I pulled these 10 steps from the migration docs so you don't learn them in production step 1 → $4 / $20 per 1M. Opus 5 was $5 / $25. cache reads dropped from $0.50 to $0.20 step 2 → 66.4% on Terminal-Bench 4.0 vs GPT-6 Astra 57.9% and Opus 5 52.3%. +14.1 points in one release, and on FrontierCode it beats Astra at default effort for 1/5 the cost step 3 → thinking can't be turned off anymore. send thinking: disabled and you get a 400. drop the field, set effort step 4 → tool_choice any and tool are gone. 400. switch to auto + strict step 5 → edit anything above a thinking block and the request dies. append only, or opt into drop_block step 6 → computer_20251124 is dead on the API. 400. move to computer_toolset_20260801 step 7 → the quiet one: default effort fell from high to medium. your agent thinks less than you set it up to and nothing tells you step 8 → hop Opus 5.5 → Sonnet 5 → Opus 5.5 and you pay 4.36 instead of 3.32. +31%, the cache dies and Sonnet can't read Opus's reasoning step 9 → change effort at the top of the request and the cache is gone. Jev sets it per message and the cache stays step 10 → switch fast - standard mid-session and it's a full cache miss. Jev picks speed once, on turn one one model, three knobs, zero 400s. that is Jev + Opus 5.5 send this to your Claude Code before you touch the model ID, then read my full Jev deep dive in the article below ↓

Carnage

16,674 views • 20 days ago

One of the best things to build right now is an action app. An action app clears your inbox, books your calendar, and files your expenses on your behalf. The AI-first version of email answers 80% of your messages and shows you the one or two a day that are too important to automate. Every app you use today works the other way around. Instagram and TikTok feed you stuff to scroll. Salesforce sits there waiting for you to type. Here's the whole setup: 1) Pick a boring workflow in a niche where people already pay to make it go away. 2) Reimagine the core UX. Don't rebuild the list of everything you owe someone, because the list is the daunting part. 3) Build on the Claude Agent SDK and other SDKs. You're putting a layer on top of it. People called them GPT wrappers, now it's agent wrappers. 4) Give people an interface for managing how the agents work. 5) Figure out the TikTok or the Reel that gets someone to install it. Step 5 is the hard part. Everyone can build the agent now, so the winner is whoever gets the reach. Facebook had a painful transition to mobile, so they bought Instagram, the mobile-first version of Facebook. Agent-first apps are the same setup. Everyone with AI bolted onto the side of an old product can't rip out their core experience, and that's the opening. Someone should start a studio that only builds these. Pick the boring workflow, let the agents run it, then go get acquired
8:35

One of the best things to build right now is an action app. An action app clears your inbox, books your calendar, and files your expenses on your behalf. The AI-first version of email answers 80% of your messages and shows you the one or two a day that are too important to automate. Every app you use today works the other way around. Instagram and TikTok feed you stuff to scroll. Salesforce sits there waiting for you to type. Here's the whole setup: 1) Pick a boring workflow in a niche where people already pay to make it go away. 2) Reimagine the core UX. Don't rebuild the list of everything you owe someone, because the list is the daunting part. 3) Build on the Claude Agent SDK and other SDKs. You're putting a layer on top of it. People called them GPT wrappers, now it's agent wrappers. 4) Give people an interface for managing how the agents work. 5) Figure out the TikTok or the Reel that gets someone to install it. Step 5 is the hard part. Everyone can build the agent now, so the winner is whoever gets the reach. Facebook had a painful transition to mobile, so they bought Instagram, the mobile-first version of Facebook. Agent-first apps are the same setup. Everyone with AI bolted onto the side of an old product can't rip out their core experience, and that's the opening. Someone should start a studio that only builds these. Pick the boring workflow, let the agents run it, then go get acquired

The Startup Ideas Podcast (SIP) 🧃

14,323 views • 1 month ago