Загрузка видео...

Не удалось загрузить видео

На главную

What's the tool stack for a marketing engineer? Here's how what I'd use: Start with GrokBot. It is connected to the X ecosystem, so give it 4 clear lanes: - Watch the competitors and report what changed - Watch customer language on X and Reddit - Watch the creators...

startupideaspod's profile picture

The Startup Ideas Podcast (SIP) 🧃

47,226 subscribers

16,495 просмотров • 9 дней назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Local AI 101: open models, Hugging Face, and businesses to build (38 min masterclass) I still think cloud AI is the default for most things, and honestly it should be, the frontier models are the strongest and easiest to use. But something shifted in the last 4-5 months. You can now run genuinely good open models directly on your own laptop, or even your phone. And once you actually try it, it changes how you think about what AI is even for. LOCAL AI, CLEARLY EXPLAINED: 1. The model is the brain doing the thinking. Gemma, Llama, Mistral, and Qwen are the main families, and each is better at different things, some at reasoning, some at coding, some small enough to run on a phone. 2. Hugging Face is the warehouse where you find them. You go there to see what each model is good at, check the license, and grab the compressed versions that run on a normal computer. 3. The software is what runs the model on your machine. Start with LM Studio if you're not technical, it feels like a normal app where you search, download, and start chatting. Ollama is the one you reach for when you want to plug a model into your own apps. 4. The workflow is the actual product you build on top of it all. That's what I'm ideating around for some businesses to create. I think local AI just made a specific kind of business way easier to start. Find an industry that: 1. Sits on sensitive data they'd never paste into ChatGPT 2. Does the same review over and over 3. Runs on software from 2003 Then build a local AI tool that does that review on their own machine, so the data never leaves the building! Take home health agencies. Nurses write visit notes all day, and if a note is missing a detail, the billing gets denied or the audit flags it. Here's how I'd start: 1. Find 5 small agencies. Offer to review a batch of their notes for them. 2. Run the notes through Gemma locally (free, private, no cloud). Read every output yourself. 3. Write down the 20 issues that keep showing up: missing vitals, vague med changes, notes that don't support the billed level. 4. That list of 20 is your checklist. The checklist is the product. 5. Turn it into a local desktop app that flags those 20 things before a note gets submitted. You just went from a service anyone could offer to a product nobody else has, and you learned exactly what to build by doing the work by hand first. Same recipe works for restoration contractors (draft the damage report on-site before the tech leaves) and wealth advisors (catch the compliance landmine in a client email before it sends). Basically the framework is sensitive data, repeated review, ancient software. I think there are tons of businesses like this! Almost none of it clicked for me until I actually started using local AI. So if you take one thing from this, go run a model on your own machine once. Also a fun thing to try with your friends. Feel free to send this to a friend. The episode is live for free on The Startup Ideas Podcast (SIP) 🧃 (thanks to Google for sponsoring today's episode and supporting local AI) I feel like local AI one of those things you need to try for it to really click. Run one model on your own machine and you'll see what I mean! I go way deeper in the full 38 minute masterclass, the models, the setup, and the businesses to build. Link below. LINK TO WATCH: OR WATCH BELOW ON X What do you think of local AI?
38:45

Local AI 101: open models, Hugging Face, and businesses to build (38 min masterclass) I still think cloud AI is the default for most things, and honestly it should be, the frontier models are the strongest and easiest to use. But something shifted in the last 4-5 months. You can now run genuinely good open models directly on your own laptop, or even your phone. And once you actually try it, it changes how you think about what AI is even for. LOCAL AI, CLEARLY EXPLAINED: 1. The model is the brain doing the thinking. Gemma, Llama, Mistral, and Qwen are the main families, and each is better at different things, some at reasoning, some at coding, some small enough to run on a phone. 2. Hugging Face is the warehouse where you find them. You go there to see what each model is good at, check the license, and grab the compressed versions that run on a normal computer. 3. The software is what runs the model on your machine. Start with LM Studio if you're not technical, it feels like a normal app where you search, download, and start chatting. Ollama is the one you reach for when you want to plug a model into your own apps. 4. The workflow is the actual product you build on top of it all. That's what I'm ideating around for some businesses to create. I think local AI just made a specific kind of business way easier to start. Find an industry that: 1. Sits on sensitive data they'd never paste into ChatGPT 2. Does the same review over and over 3. Runs on software from 2003 Then build a local AI tool that does that review on their own machine, so the data never leaves the building! Take home health agencies. Nurses write visit notes all day, and if a note is missing a detail, the billing gets denied or the audit flags it. Here's how I'd start: 1. Find 5 small agencies. Offer to review a batch of their notes for them. 2. Run the notes through Gemma locally (free, private, no cloud). Read every output yourself. 3. Write down the 20 issues that keep showing up: missing vitals, vague med changes, notes that don't support the billed level. 4. That list of 20 is your checklist. The checklist is the product. 5. Turn it into a local desktop app that flags those 20 things before a note gets submitted. You just went from a service anyone could offer to a product nobody else has, and you learned exactly what to build by doing the work by hand first. Same recipe works for restoration contractors (draft the damage report on-site before the tech leaves) and wealth advisors (catch the compliance landmine in a client email before it sends). Basically the framework is sensitive data, repeated review, ancient software. I think there are tons of businesses like this! Almost none of it clicked for me until I actually started using local AI. So if you take one thing from this, go run a model on your own machine once. Also a fun thing to try with your friends. Feel free to send this to a friend. The episode is live for free on The Startup Ideas Podcast (SIP) 🧃 (thanks to Google for sponsoring today's episode and supporting local AI) I feel like local AI one of those things you need to try for it to really click. Run one model on your own machine and you'll see what I mean! I go way deeper in the full 38 minute masterclass, the models, the setup, and the businesses to build. Link below. LINK TO WATCH: OR WATCH BELOW ON X What do you think of local AI?

GREG ISENBERG

26,928 просмотров • 1 день назад

THIS GUY CONNECTED HIS AI AGENTS TO HIS OBSIDIAN AND BUILT A BRAIN THAT LEARNS ON ITS OWN. HERE'S HOW TO BUILD IT Obsidian is just markdown files sitting in a folder. That turns out to be the perfect memory for an AI agent, because an agent can read and write those files directly. He wired his agents into the vault so they pull context from it, do the work, and write what they learned back. The notes aren't the point. The loop is, and it gets sharper every cycle How to build it: 1. Point an agent at your vault. The fastest way, no plugins, no API keys: open a terminal and run npx obsidian-mcp /path/to/your/vault. That exposes your Obsidian folder to Claude as a tool it can read, search, and write to. Add it to your Claude Code or Cowork config and restart 2. Confirm it can see the brain. Ask it: "list the notes in my vault and summarize what's in them." If it reads them back, the connection is live. Now it starts every task with everything the vault already holds instead of from zero 3. Give each agent one job and a write-back rule. Tell it: "research this, then save what you found as a new note in /brain with links to related notes." One agent researches, one summarizes, one plans. Each writes its output back into the vault 4. Close the loop. Add one line to every agent's instructions: "read /brain before starting, write your result back when done." Now each task leaves the vault richer, and the next run reads that before it works. It compounds instead of resetting 5. You only steer. Review what the brain produces, point it at the next thing. The agents handle the reading, writing, and connecting The edge isn't better notes. It's a brain that feeds itself, so the work gets sharper every cycle instead of starting over Bookmark this
0:24

THIS GUY CONNECTED HIS AI AGENTS TO HIS OBSIDIAN AND BUILT A BRAIN THAT LEARNS ON ITS OWN. HERE'S HOW TO BUILD IT Obsidian is just markdown files sitting in a folder. That turns out to be the perfect memory for an AI agent, because an agent can read and write those files directly. He wired his agents into the vault so they pull context from it, do the work, and write what they learned back. The notes aren't the point. The loop is, and it gets sharper every cycle How to build it: 1. Point an agent at your vault. The fastest way, no plugins, no API keys: open a terminal and run npx obsidian-mcp /path/to/your/vault. That exposes your Obsidian folder to Claude as a tool it can read, search, and write to. Add it to your Claude Code or Cowork config and restart 2. Confirm it can see the brain. Ask it: "list the notes in my vault and summarize what's in them." If it reads them back, the connection is live. Now it starts every task with everything the vault already holds instead of from zero 3. Give each agent one job and a write-back rule. Tell it: "research this, then save what you found as a new note in /brain with links to related notes." One agent researches, one summarizes, one plans. Each writes its output back into the vault 4. Close the loop. Add one line to every agent's instructions: "read /brain before starting, write your result back when done." Now each task leaves the vault richer, and the next run reads that before it works. It compounds instead of resetting 5. You only steer. Review what the brain produces, point it at the next thing. The agents handle the reading, writing, and connecting The edge isn't better notes. It's a brain that feeds itself, so the work gets sharper every cycle instead of starting over Bookmark this

Yarchi

58,186 просмотров • 3 месяцев назад

What does it take to make AI models think faster and more accurately? Watch for an insight into the groundbreaking work of a Huawei AI engineer, who is helping the company keep pace with the compute surge.
1:00

What does it take to make AI models think faster and more accurately? Watch for an insight into the groundbreaking work of a Huawei AI engineer, who is helping the company keep pace with the compute surge.

Huawei

8,062,380 просмотров • 22 дней назад

The "marketing engineer" is the NEW forward deployed engineer, and I think the BEST ones will make $1M a year! A forward deployed engineer embeds with your team and uses AI to build the workflows The marketing engineer does that BUT for growth, they build AI agents that find your customers, write your outbound, test your ads, and get smarter every week. The most valuable marketer changes with EVERY major tech wave: 1. Traditional marketer: make people care with story (print, radio etc) 2. Digital marketer: the marketer who owned new digital channels (SEO, PPC) 3. Growth hacker: the growth marketer who lived in loops, PLG and retention 4. Marketing engineer: the marketer who builds AI agents that run the whole system. The type of agents a marketing engineer would create: 1. The customer language agent. It pulls your Gong call transcripts, your Intercom tickets, and your G2 reviews every week, extracts the exact words customers use to describe the problem, and drops a memo ranked by how often each phrase shows up. 2. The buying trigger agent. It watches for signals that someone's ready, a company posting a job for the role you sell to, a funding announcement, a competitor getting torched in a review, then enriches the contact through Apollo and drafts the outbound tied to that exact trigger the second it fires. 3. The SEO gap agent. It pulls keyword gaps from Ahrefs, checks who's already ranking on page one, reads the top three results, then writes a better post with the founder's actual take baked in, plus the meta title and internal links, and drops it in for approval. 4. The creative testing agent. It takes one offer, generates 100 ad variations across three different angles with Nano Banana and your copy model, pushes them live through the Meta API, and kills anything under a 1% CTR on its own so only the winners keep spending. AND MANY MORE. If you're a marketer, this is how you stop being replaceable. If you're a founder, this is how you grow your company efficiently. AI agents are here and marketers are about to have a field day!! I explain everything (tools, agents, 30d plan) on today's episode of The Startup Ideas Podcast (SIP) 🧃 Watch: --> Marketing engineers are here. Whatever we end up calling them, the top 1% of them are going to make an absurd amount of money.
35:18

The "marketing engineer" is the NEW forward deployed engineer, and I think the BEST ones will make $1M a year! A forward deployed engineer embeds with your team and uses AI to build the workflows The marketing engineer does that BUT for growth, they build AI agents that find your customers, write your outbound, test your ads, and get smarter every week. The most valuable marketer changes with EVERY major tech wave: 1. Traditional marketer: make people care with story (print, radio etc) 2. Digital marketer: the marketer who owned new digital channels (SEO, PPC) 3. Growth hacker: the growth marketer who lived in loops, PLG and retention 4. Marketing engineer: the marketer who builds AI agents that run the whole system. The type of agents a marketing engineer would create: 1. The customer language agent. It pulls your Gong call transcripts, your Intercom tickets, and your G2 reviews every week, extracts the exact words customers use to describe the problem, and drops a memo ranked by how often each phrase shows up. 2. The buying trigger agent. It watches for signals that someone's ready, a company posting a job for the role you sell to, a funding announcement, a competitor getting torched in a review, then enriches the contact through Apollo and drafts the outbound tied to that exact trigger the second it fires. 3. The SEO gap agent. It pulls keyword gaps from Ahrefs, checks who's already ranking on page one, reads the top three results, then writes a better post with the founder's actual take baked in, plus the meta title and internal links, and drops it in for approval. 4. The creative testing agent. It takes one offer, generates 100 ad variations across three different angles with Nano Banana and your copy model, pushes them live through the Meta API, and kills anything under a 1% CTR on its own so only the winners keep spending. AND MANY MORE. If you're a marketer, this is how you stop being replaceable. If you're a founder, this is how you grow your company efficiently. AI agents are here and marketers are about to have a field day!! I explain everything (tools, agents, 30d plan) on today's episode of The Startup Ideas Podcast (SIP) 🧃 Watch: --> Marketing engineers are here. Whatever we end up calling them, the top 1% of them are going to make an absurd amount of money.

GREG ISENBERG

233,409 просмотров • 9 дней назад

Andrej Karpathy spent 2h showing how he actually uses AI day to day he's a co-founder of OpenAI and led AI at Tesla, so when he shows how he works, it’s worth watching and the whole session is just him telling the machine what he wants in simple terms, like he's briefing a coworker watch what's actually happening the entire time: > he describes the task in normal words > it goes off and does the work > he glances at the result and nudges it with one more sentence that's the whole skill, and you've had it since you learned to talk the only gap between that and a worker that runs on its own is handing that sentence a schedule and the tools to act check his work, then build the version that keeps working when you stop
2:11:11

Andrej Karpathy spent 2h showing how he actually uses AI day to day he's a co-founder of OpenAI and led AI at Tesla, so when he shows how he works, it’s worth watching and the whole session is just him telling the machine what he wants in simple terms, like he's briefing a coworker watch what's actually happening the entire time: > he describes the task in normal words > it goes off and does the work > he glances at the result and nudges it with one more sentence that's the whole skill, and you've had it since you learned to talk the only gap between that and a worker that runs on its own is handing that sentence a schedule and the tools to act check his work, then build the version that keeps working when you stop

Chrome

1,809,883 просмотров • 3 месяцев назад

My whole company runs on skills I bundled into a GitHub repo. Here's how to build the same thing for your team: - Put your skills in a GitHub repo - Get Claude to write the couple of JSON files that make it readable as a plugin - In Claude Code, type /plugin, go to marketplaces, add marketplace, paste your repo URL - Split the plugins by department, so your copywriter isn't installing the finance skills It works in Codex too. I've got the same team skills installed there and I use them the same way. Now when I build something like my Notion formatting skill, I add it to the repo and it lands on everyone's Claude and Codex. If someone wrecks a skill, we roll it back to an older version. The key is enabling auto-update, so when someone improves a skill everyone benefits. I don't have a clue how the JSON works. I got Claude to set the whole thing up for me. Start with one skill and before you know it your whole team will be AI pilled.
5:18

My whole company runs on skills I bundled into a GitHub repo. Here's how to build the same thing for your team: - Put your skills in a GitHub repo - Get Claude to write the couple of JSON files that make it readable as a plugin - In Claude Code, type /plugin, go to marketplaces, add marketplace, paste your repo URL - Split the plugins by department, so your copywriter isn't installing the finance skills It works in Codex too. I've got the same team skills installed there and I use them the same way. Now when I build something like my Notion formatting skill, I add it to the repo and it lands on everyone's Claude and Codex. If someone wrecks a skill, we roll it back to an older version. The key is enabling auto-update, so when someone improves a skill everyone benefits. I don't have a clue how the JSON works. I got Claude to set the whole thing up for me. Start with one skill and before you know it your whole team will be AI pilled.

The Startup Ideas Podcast (SIP) 🧃

19,532 просмотров • 20 дней назад

One of my favorite ways to use Claude Code is to create a ticket Here is why I love it: Claude Code is very good at doing the work, but it must know when it's "done." A ticket gives Claude Code a clear signpost for when to stop Here's how to write it in 4 parts: 1) Give it a job: - "Add a wait list form to the landing page." 2) Give it a proper scope: - Ask for a name, an email, and a company. 3) Tell it the UX you want: - Show a simple success message after a person sends the form. 4) Tell the constraints it needs to follow: - Keep the design the same as the current brand identity. Then add an instruction: make the change small enough to review in the diff view, run the checks, open the app in the desktop preview, and tell me what still needs human review. The catch: a ticket that is not clear makes Claude guess. "Make the app better." "Add AI." "Build the whole thing." You get a diff too big to read, so you cannot tell if it is right. Keep the ticket small enough so that Claude knows exactly what to do
4:07

One of my favorite ways to use Claude Code is to create a ticket Here is why I love it: Claude Code is very good at doing the work, but it must know when it's "done." A ticket gives Claude Code a clear signpost for when to stop Here's how to write it in 4 parts: 1) Give it a job: - "Add a wait list form to the landing page." 2) Give it a proper scope: - Ask for a name, an email, and a company. 3) Tell it the UX you want: - Show a simple success message after a person sends the form. 4) Tell the constraints it needs to follow: - Keep the design the same as the current brand identity. Then add an instruction: make the change small enough to review in the diff view, run the checks, open the app in the desktop preview, and tell me what still needs human review. The catch: a ticket that is not clear makes Claude guess. "Make the app better." "Add AI." "Build the whole thing." You get a diff too big to read, so you cannot tell if it is right. Keep the ticket small enough so that Claude knows exactly what to do

The Startup Ideas Podcast (SIP) 🧃

10,471 просмотров • 23 дней назад

Introducing Magnitude: your actually local agent 100% private and offline. No token costs, no API keys. Open source. Today's agents are local. The model isn't. Every prompt, every file, every secret gets sent straight to Anthropic and OpenAI. Magnitude is built around local models and runs the whole stack itself. The inference engine is part of the agent, so the models run inside it, right on your computer. It lives in your terminal, and setup is one command. Magnitude profiles your hardware and shows you which models fit, with the trade-offs between quality, speed, and memory. Pick one and start working. No painful config or server to babysit. Out of the box, it can use your shell, edit files, and run scripts. Add skills and it can work with Excel, PowerPoint, PDFs or Chrome. Use it for everyday work: - Analyze sensitive data - Manage private notes - Review code and logs - Search and organize files - Build docs or slides npm i -g @magnitudedev/cli GitHub:
0:35

Introducing Magnitude: your actually local agent 100% private and offline. No token costs, no API keys. Open source. Today's agents are local. The model isn't. Every prompt, every file, every secret gets sent straight to Anthropic and OpenAI. Magnitude is built around local models and runs the whole stack itself. The inference engine is part of the agent, so the models run inside it, right on your computer. It lives in your terminal, and setup is one command. Magnitude profiles your hardware and shows you which models fit, with the trade-offs between quality, speed, and memory. Pick one and start working. No painful config or server to babysit. Out of the box, it can use your shell, edit files, and run scripts. Add skills and it can work with Excel, PowerPoint, PDFs or Chrome. Use it for everyday work: - Analyze sensitive data - Manage private notes - Review code and logs - Search and organize files - Build docs or slides npm i -g @magnitudedev/cli GitHub:

Tom Greenwald

395,703 просмотров • 1 месяц назад

Ready to build the future of stable private on-device AI? 🧠 Our latest tutorial shows you how to build a sovereign mobile app in minutes using the QVAC SDK and Expo. Start from a blank template and deploy in minutes a local Llama 3.2 inference running directly on your own devices. What you’ll learn: Modular Setup: Use the QVAC CLI to tree-shake and keep your mobile bundle lean. Local-First Flow: Initialize the SDK, download weights, and run high-speed inference without a cloud uplink. Cross-Platform Power: See the smoke test in action on a physical Samsung S25. No rented clouds. No API keys. Build local, on-device, unstoppable intelligence in your pocket. Watch the full guide and start building:
5:37

Ready to build the future of stable private on-device AI? 🧠 Our latest tutorial shows you how to build a sovereign mobile app in minutes using the QVAC SDK and Expo. Start from a blank template and deploy in minutes a local Llama 3.2 inference running directly on your own devices. What you’ll learn: Modular Setup: Use the QVAC CLI to tree-shake and keep your mobile bundle lean. Local-First Flow: Initialize the SDK, download weights, and run high-speed inference without a cloud uplink. Cross-Platform Power: See the smoke test in action on a physical Samsung S25. No rented clouds. No API keys. Build local, on-device, unstoppable intelligence in your pocket. Watch the full guide and start building:

QVAC

4,081,211 просмотров • 4 месяцев назад

HERMES CAN NOW RUN FOR FREE FOREVER A new Gemma 4 update just removed the biggest problem with local AI agents. The Upgrade: → Gemma 4 runs around 90% faster on Apple Silicon using Ollama + MLX → Plug it into Hermes with a few terminal commands → Run private agent workflows without subscriptions, token limits, or API bills What It Can Actually Do: ✓ Build and preview apps locally ✓ Use /learn to turn tutorials into permanent skills ✓ Read files, analyse landing pages, summarise emails, and track AI news ✓ Run loops until a task is complete instead of waiting for another prompt ✓ Launch multiple Gemma 4 subagents to work on separate tasks in parallel No Apple Silicon? ✔ OpenRouter offers the 31B model through a free API The real win is not faster chat. It is having Hermes build, retry, check its work, and operate in the background without charging you for every loop.
10:36

HERMES CAN NOW RUN FOR FREE FOREVER A new Gemma 4 update just removed the biggest problem with local AI agents. The Upgrade: → Gemma 4 runs around 90% faster on Apple Silicon using Ollama + MLX → Plug it into Hermes with a few terminal commands → Run private agent workflows without subscriptions, token limits, or API bills What It Can Actually Do: ✓ Build and preview apps locally ✓ Use /learn to turn tutorials into permanent skills ✓ Read files, analyse landing pages, summarise emails, and track AI news ✓ Run loops until a task is complete instead of waiting for another prompt ✓ Launch multiple Gemma 4 subagents to work on separate tasks in parallel No Apple Silicon? ✔ OpenRouter offers the 31B model through a free API The real win is not faster chat. It is having Hermes build, retry, check its work, and operate in the background without charging you for every loop.

Julian Goldie SEO

11,363 просмотров • 2 месяцев назад

SpaceXAI engineer (ex-Cursor): "GrokBot is the most powerful agentic tool, but people use only 10% of it's its potential. right now 20+ Grokbot's reads my X bookmarks, writes the prompts, hire a Cursor cloud agents, ships the demo and hands me a live preview URL. I just say yes" in a 16-minute episode, a SpaceXAI engineer reveal how to build 4 levels of GrokBot from "easy" to "god mode" one bot → bots calling bots → a CRM of 2,500 people → agents that write the prompt and ship the code after this episode I asked for a refund on my $500 agentic AI course watch today, then read the article below on how to build a team of GrokBot agents from scratch
36:32

SpaceXAI engineer (ex-Cursor): "GrokBot is the most powerful agentic tool, but people use only 10% of it's its potential. right now 20+ Grokbot's reads my X bookmarks, writes the prompts, hire a Cursor cloud agents, ships the demo and hands me a live preview URL. I just say yes" in a 16-minute episode, a SpaceXAI engineer reveal how to build 4 levels of GrokBot from "easy" to "god mode" one bot → bots calling bots → a CRM of 2,500 people → agents that write the prompt and ship the code after this episode I asked for a refund on my $500 agentic AI course watch today, then read the article below on how to build a team of GrokBot agents from scratch

Codez

18,188 просмотров • 10 дней назад

Anthropic just dropped a free 1-hour AI engineering workshop On how their engineers actually build with AI in 2026: 00:26 - Why prompting is becoming the old way 16:57 - The loop that builds, tests, and fixes its own work 25:27 - How one engineer can do the work of a 5-person team 36:25 - Build your first AI loop from scratch This 1-hour watch is worth more than most paid AI engineering courses Writing code is becoming the easy part Building the loop is the new skill Bookmark and watch it today Then build your own loop with the guide below
1:11:54

Anthropic just dropped a free 1-hour AI engineering workshop On how their engineers actually build with AI in 2026: 00:26 - Why prompting is becoming the old way 16:57 - The loop that builds, tests, and fixes its own work 25:27 - How one engineer can do the work of a 5-person team 36:25 - Build your first AI loop from scratch This 1-hour watch is worth more than most paid AI engineering courses Writing code is becoming the easy part Building the loop is the new skill Bookmark and watch it today Then build your own loop with the guide below

Lunar

15,125 просмотров • 1 месяц назад

You can spin up an AI workforce with one prompt. Here's why it works: it pulls the context the AI needs to lay the foundation. Give it: - your role - team size - the goals you want to hit. Then say "interview me." After that, build it out in four steps: 1) Start with 1 agent. - What does it feel like to work with one? 2) Make that agent proactive. - Now it's doing things on your behalf before you ask. 3) Expand to 2 agents. - One directs the other, or routes to it. - This is where you learn how context gets passed. 4) Scale the workforce. - A mission control where you watch it all move around. - Now you see what has to run in parallel, which agent never needed that tool, which one runs fine on a smaller model. - Not everything needs Opus. Sub-agents are haiku and sonnet. The catch: the prompt gets you agents with markdown files doing some stuff. Getting to the 90% plus level is iteration, and that part is specific to each person. Nobody can hand it to you. Start with one agent. Spin up one agent this week. The rest is discovery.
3:10

You can spin up an AI workforce with one prompt. Here's why it works: it pulls the context the AI needs to lay the foundation. Give it: - your role - team size - the goals you want to hit. Then say "interview me." After that, build it out in four steps: 1) Start with 1 agent. - What does it feel like to work with one? 2) Make that agent proactive. - Now it's doing things on your behalf before you ask. 3) Expand to 2 agents. - One directs the other, or routes to it. - This is where you learn how context gets passed. 4) Scale the workforce. - A mission control where you watch it all move around. - Now you see what has to run in parallel, which agent never needed that tool, which one runs fine on a smaller model. - Not everything needs Opus. Sub-agents are haiku and sonnet. The catch: the prompt gets you agents with markdown files doing some stuff. Getting to the 90% plus level is iteration, and that part is specific to each person. Nobody can hand it to you. Start with one agent. Spin up one agent this week. The rest is discovery.

The Startup Ideas Podcast (SIP) 🧃

18,006 просмотров • 27 дней назад

I just vibe-coded a Meta ad research app in Claude Code 🤯 One keyword search → winning ads analyzed, creative briefs generated, trends mapped, and 10 ad variations written for your brand. All inside Claude Code. Perfect for DTC brands and agencies who are still doing competitor ad research manually inside the Meta Ad Library. If you're clicking through ads one by one, watching videos, taking notes in a Google Doc, trying to reverse-engineer what's working, and then rewriting briefs from scratch every time... This app eliminates the entire loop: → Search and filter winning Facebook ads by niche, country, language, and performance tier → Watch video ads inline without leaving the app → Fetch top-performing ads and generate a full creative brief tailored to your brand → Run a trend radar across 100+ ads to see which formats, CTAs, and landing pages are dominating your niche → Pick any winning video ad and Gemini watches it, reverse-engineers the creative DNA, and generates 10 new ad variations for your product No scrolling the Ad Library for hours. No manual note-taking. No rewriting briefs from scratch. What you get: → Creative briefs generated from real winning ads in your niche → Trend analysis with format distribution, top CTAs, landing page intel, and AI insights → 10 brand-specific ad variations from any winning video — with hooks, scripts, and Nano Banana prompts → A full app you host in Replit and customize for your team Built 100% in Claude Code with the gethookd.ai API + Gemini. I put together a free playbook with every prompt I used to build this app from scratch, so you can build it yourself. Want it for free? > Like this post > Comment "META" And I'll send it over (must be following so I can DM)
0:20

I just vibe-coded a Meta ad research app in Claude Code 🤯 One keyword search → winning ads analyzed, creative briefs generated, trends mapped, and 10 ad variations written for your brand. All inside Claude Code. Perfect for DTC brands and agencies who are still doing competitor ad research manually inside the Meta Ad Library. If you're clicking through ads one by one, watching videos, taking notes in a Google Doc, trying to reverse-engineer what's working, and then rewriting briefs from scratch every time... This app eliminates the entire loop: → Search and filter winning Facebook ads by niche, country, language, and performance tier → Watch video ads inline without leaving the app → Fetch top-performing ads and generate a full creative brief tailored to your brand → Run a trend radar across 100+ ads to see which formats, CTAs, and landing pages are dominating your niche → Pick any winning video ad and Gemini watches it, reverse-engineers the creative DNA, and generates 10 new ad variations for your product No scrolling the Ad Library for hours. No manual note-taking. No rewriting briefs from scratch. What you get: → Creative briefs generated from real winning ads in your niche → Trend analysis with format distribution, top CTAs, landing page intel, and AI insights → 10 brand-specific ad variations from any winning video — with hooks, scripts, and Nano Banana prompts → A full app you host in Replit and customize for your team Built 100% in Claude Code with the gethookd.ai API + Gemini. I put together a free playbook with every prompt I used to build this app from scratch, so you can build it yourself. Want it for free? > Like this post > Comment "META" And I'll send it over (must be following so I can DM)

Mike Futia

53,558 просмотров • 5 месяцев назад

Boris Cherny, the engineer who built Claude Code: "I don't talk to an agent anymore. I talk to a loop" a prompt is one instruction you babysit. a loop is a goal the AI works toward on its own: > it plans > does the work > checks itself > fixes what's weak > and repeats until it's done you step out, the work keeps going that's the shift the best engineers quietly moved to. most people hear a line like that and have no idea what it means in practice this breaks it down without the hype: what a loop actually is, when it's worth building, when it's a money trap, and how to run one yourself today watch his clip, then read the article below
0:40

Boris Cherny, the engineer who built Claude Code: "I don't talk to an agent anymore. I talk to a loop" a prompt is one instruction you babysit. a loop is a goal the AI works toward on its own: > it plans > does the work > checks itself > fixes what's weak > and repeats until it's done you step out, the work keeps going that's the shift the best engineers quietly moved to. most people hear a line like that and have no idea what it means in practice this breaks it down without the hype: what a loop actually is, when it's worth building, when it's a money trap, and how to run one yourself today watch his clip, then read the article below

Dep

20,934 просмотров • 2 месяцев назад

RIP local SEO agencies. claude just one-shot your entire job and it took one prompt. here's the full breakdown: the workflow is stupidly simple. jacky didn't even look at the client's site. he transcribed the previous episode, exported the local SEO ebook, gave both to claude and said fix this. out came a massive document: hero section changes, schema gaps, missing pages, internal linking fixes, and a full build order of what to ship first. next step is feeding this call's transcript back in, getting the final mockup, and handing it to the team to one-shot. claude suggested pages that didn't exist and they're genuinely smart. practitioner pages so the clinic ranks when someone googles their physio by name. neighborhood pages like "physiotherapy in killarney, 7 minutes from our victoria drive clinic" to capture every surrounding area search. pricing FAQs. the neighborhood play works because people search their specific scenario now, and AI search made long tail queries explode. the one warning: don't blast 60 boilerplate pages at once. tristan audited a member's site this week that got deindexed for exactly that. the guy one-shotted the whole site with templated content and google sandboxed it, and it's way harder to come back out of the sandbox now. rewrite every page, phase them in, be surgical. the interlinking structure is the actual sauce. service pages stay location neutral (no city terms in the H1 or you cannibalize), then link down to their east vancouver and mount pleasant versions. each location page links back to its respective service area pages. one to one targeting in every area. tristan's words: once your authority and reviews build up on top of that structure, it's game over. RIP wordpress too while we're at it. the only clients still on it are the ones whose teams can't leave. everything new runs out of github and cloudflare, changes ship in one sentence to claude ("put these six service pages in the menu nav and deploy"), and the sites look infinitely better. one advise member keeps a stack of HTML templates and just runs one up every time he finds a rank and rent. that's the stack now. the meta lesson: tristan's buddy in singapore watched the last episode, took the transcript, and had his codex agent running all day shipping every change to his new jersey site. citations in, indexed, tier 2s queued. he's probably gonna make it. the era of watching a youtube video and having AI implement it before the next episode drops is fully here. next up: a live rank and rent case study, national level, built in public. niche suggestions wanted. full claude local SEO takeover with Tristan Zheng | SEO Consultant watch/listen ↓
27:19

RIP local SEO agencies. claude just one-shot your entire job and it took one prompt. here's the full breakdown: the workflow is stupidly simple. jacky didn't even look at the client's site. he transcribed the previous episode, exported the local SEO ebook, gave both to claude and said fix this. out came a massive document: hero section changes, schema gaps, missing pages, internal linking fixes, and a full build order of what to ship first. next step is feeding this call's transcript back in, getting the final mockup, and handing it to the team to one-shot. claude suggested pages that didn't exist and they're genuinely smart. practitioner pages so the clinic ranks when someone googles their physio by name. neighborhood pages like "physiotherapy in killarney, 7 minutes from our victoria drive clinic" to capture every surrounding area search. pricing FAQs. the neighborhood play works because people search their specific scenario now, and AI search made long tail queries explode. the one warning: don't blast 60 boilerplate pages at once. tristan audited a member's site this week that got deindexed for exactly that. the guy one-shotted the whole site with templated content and google sandboxed it, and it's way harder to come back out of the sandbox now. rewrite every page, phase them in, be surgical. the interlinking structure is the actual sauce. service pages stay location neutral (no city terms in the H1 or you cannibalize), then link down to their east vancouver and mount pleasant versions. each location page links back to its respective service area pages. one to one targeting in every area. tristan's words: once your authority and reviews build up on top of that structure, it's game over. RIP wordpress too while we're at it. the only clients still on it are the ones whose teams can't leave. everything new runs out of github and cloudflare, changes ship in one sentence to claude ("put these six service pages in the menu nav and deploy"), and the sites look infinitely better. one advise member keeps a stack of HTML templates and just runs one up every time he finds a rank and rent. that's the stack now. the meta lesson: tristan's buddy in singapore watched the last episode, took the transcript, and had his codex agent running all day shipping every change to his new jersey site. citations in, indexed, tier 2s queued. he's probably gonna make it. the era of watching a youtube video and having AI implement it before the next episode drops is fully here. next up: a live rank and rent case study, national level, built in public. niche suggestions wanted. full claude local SEO takeover with Tristan Zheng | SEO Consultant watch/listen ↓

Jacky Chou (buying online businesses up to $1m)

24,078 просмотров • 1 месяц назад

A DEVELOPER CONNECTED CLAUDE CODE TO OBSIDIAN SO HIS AI AGENT WOULD STOP FORGETTING THE PROJECT EVERY MORNING. Every coding session used to start the same way. Claude would understand the repo, fix the bug, explain the architecture, and then the moment the session ended, all of that context disappeared. Same codebase. Same decisions. Same architecture. Same mistakes repeated again. So he added a memory layer. Instead of treating Claude Code like a smart terminal, he connected it to a local Obsidian vault through MCP. Now Claude can read the repo, open the vault, create notes, link concepts, and write important decisions back into the system. When it studies the codebase, it does not just answer once and forget. It creates notes for the major services, maps how the architecture works, links auth to the database, connects APIs to storage, and records why certain migrations or design choices exist. Obsidian becomes the project graph. Now when he asks why something was built a certain way, Claude does not guess from the current prompt. It reads the decision notes. When he starts a new branch, Claude checks the active context file. When the work is done, it updates what changed, what is blocked, and what the next agent needs to know before touching the repo. That is the real loop: read context, write code, capture decisions, update memory. Most people are still using AI coding tools like disposable chat windows. Ask, patch, close, forget. This setup turns Claude Code into infrastructure. The repo gets a memory layer that survives every session, and multiple AI agents can work from the same project map without stepping on each other. The unlock is not better prompting. The unlock is giving the agent somewhere to remember what it already learned.
4:19

A DEVELOPER CONNECTED CLAUDE CODE TO OBSIDIAN SO HIS AI AGENT WOULD STOP FORGETTING THE PROJECT EVERY MORNING. Every coding session used to start the same way. Claude would understand the repo, fix the bug, explain the architecture, and then the moment the session ended, all of that context disappeared. Same codebase. Same decisions. Same architecture. Same mistakes repeated again. So he added a memory layer. Instead of treating Claude Code like a smart terminal, he connected it to a local Obsidian vault through MCP. Now Claude can read the repo, open the vault, create notes, link concepts, and write important decisions back into the system. When it studies the codebase, it does not just answer once and forget. It creates notes for the major services, maps how the architecture works, links auth to the database, connects APIs to storage, and records why certain migrations or design choices exist. Obsidian becomes the project graph. Now when he asks why something was built a certain way, Claude does not guess from the current prompt. It reads the decision notes. When he starts a new branch, Claude checks the active context file. When the work is done, it updates what changed, what is blocked, and what the next agent needs to know before touching the repo. That is the real loop: read context, write code, capture decisions, update memory. Most people are still using AI coding tools like disposable chat windows. Ask, patch, close, forget. This setup turns Claude Code into infrastructure. The repo gets a memory layer that survives every session, and multiple AI agents can work from the same project map without stepping on each other. The unlock is not better prompting. The unlock is giving the agent somewhere to remember what it already learned.

DegenCalls

20,124 просмотров • 2 месяцев назад

MOST AI AGENTS RESET TO ZERO EVERY TIME YOU OPEN A NEW SESSION. Hermes Agent was built to do the opposite. Every task compounds. Every workflow becomes reusable. Every session teaches the next one. That is the core thesis behind Hermes, and it completely changes how AI agents work. Unlike most AI agents built by software tool companies, Hermes was created by the researchers training the Hermes model family itself. Meaning: The same system used internally to generate training data is now open source and available to everyone. One install gives you access to: Claude. GPT. Grok. Local models via Ollama. Telegram, Discord, Slack, iMessage integrations. Custom skills. Persistent memory. CLI control. But the real unlock is the compounding loop: You give Hermes a task. It records every tool call, decision, and result. Then it converts successful workflows into reusable skills automatically. A Hermes setup running for 3 months becomes fundamentally smarter than a fresh install. That’s the big idea. One of the coolest parts: You literally watch the agent evolve in real time from a blank install into a personalized AI operator. Instead of watching Netflix today, spend 28 mins with this. Follow Dami-Defi for more AI systems, agent workflows, and infrastructure alpha.
0:48

MOST AI AGENTS RESET TO ZERO EVERY TIME YOU OPEN A NEW SESSION. Hermes Agent was built to do the opposite. Every task compounds. Every workflow becomes reusable. Every session teaches the next one. That is the core thesis behind Hermes, and it completely changes how AI agents work. Unlike most AI agents built by software tool companies, Hermes was created by the researchers training the Hermes model family itself. Meaning: The same system used internally to generate training data is now open source and available to everyone. One install gives you access to: Claude. GPT. Grok. Local models via Ollama. Telegram, Discord, Slack, iMessage integrations. Custom skills. Persistent memory. CLI control. But the real unlock is the compounding loop: You give Hermes a task. It records every tool call, decision, and result. Then it converts successful workflows into reusable skills automatically. A Hermes setup running for 3 months becomes fundamentally smarter than a fresh install. That’s the big idea. One of the coolest parts: You literally watch the agent evolve in real time from a blank install into a personalized AI operator. Instead of watching Netflix today, spend 28 mins with this. Follow Dami-Defi for more AI systems, agent workflows, and infrastructure alpha.

Dami-Defi

11,685 просмотров • 3 месяцев назад

watch and let this sit for a second anon, a 35B AI model, building a complete multi-file game by itself on hermes agent. no cloud or api. not one line of code written by a human. running on a DGX spark in 2026. because not long ago this was science fiction. this is Ornith, a new open agentic coding model, running near lossless on a single DGX Spark, a 128GB box that sits on your desk. i hand it one spec and step back. it reasons about the architecture, creates the project structure, writes every module one by one, and here's the part that got me, it checks its own work as it goes. makes a call, reads the result, decides if it's right, corrects itself before moving on. no steering from me. this is what people keep underestimating about local AI. they hear "local" and think weaker, slower, a toy. watch the clip. a 35B at near full precision, planning and self correcting on a single desktop. the open models got genuinely good at agentic work, not benchmark good, actually builds the thing good. sped 5x so you can watch the whole build start to finish. this is where local AI actually is now. not coming soon. here.
2:29

watch and let this sit for a second anon, a 35B AI model, building a complete multi-file game by itself on hermes agent. no cloud or api. not one line of code written by a human. running on a DGX spark in 2026. because not long ago this was science fiction. this is Ornith, a new open agentic coding model, running near lossless on a single DGX Spark, a 128GB box that sits on your desk. i hand it one spec and step back. it reasons about the architecture, creates the project structure, writes every module one by one, and here's the part that got me, it checks its own work as it goes. makes a call, reads the result, decides if it's right, corrects itself before moving on. no steering from me. this is what people keep underestimating about local AI. they hear "local" and think weaker, slower, a toy. watch the clip. a 35B at near full precision, planning and self correcting on a single desktop. the open models got genuinely good at agentic work, not benchmark good, actually builds the thing good. sped 5x so you can watch the whole build start to finish. this is where local AI actually is now. not coming soon. here.

Sudo su

30,299 просмотров • 2 месяцев назад