The Startup Ideas Podcast

I run this episode solo. I explain local AI in plain terms: the model runs on hardware I control, and a cloud model runs somewhere else. I map the four pieces of the local AI landscape — the model, the warehouse, the software, and the workflow — and I define the words that beginners meet first: parameters, tokens, context window, quantization, and GGUF. I walk through the Google open model stack (Gemma 4, Google AI Edge, LiteRT-LM, AI Edge Gallery), compare the other open model families, and show three ways to run a model today. I close with a first workflow you can copy and three startup ideas that use local AI as the wedge.

And a special thank you to Google for supporting the podcast.

Timestamps

00:00 – Intro

01:35 – The Open Model the Landscape

03:09 – Vocab Decoder

06:48 – Google Gemma Clearly Explained

10:29 – Other Open Model Families

14:20 – Path 1: Run Gemma in LM Studio

18:17 – Path 2: Ollama

20:15 – Path 3: Google AI Edge

21:07 – Hardware Cheat Sheet

21:52 – First Workflow to Build

22:47 – Workflows Before Fine-Tuning

25:06 – Local vs Cloud vs Hybrid Eval

26:33 – Framework for Local AI Startup Ideas

27:22 – Startup Idea 1: Home Health QA Reviewer

29:24 – Startup Idea 2: Offline Field Report Copilot

32:10 – Startup Idea 3: Pre-Send Reviewer for Professional Services

34:47 – Build Your Local AI Lab

37:55 – Closing Thoughts

Key Points

• Ask whether the model is good enough for the job, and the business opportunities become clear.
• Local AI has four pieces: the model, the warehouse (Hugging Face), the software (LM Studio or Ollama), and the workflow you build around them.
• Gemma 4 E4B is my practical starting point; E2B fits phones and older machines.
• Hybrid architecture wins: local does the private first pass, cloud does the heavy reasoning, and a human approves anything important.
• Start with one repeated workflow — one folder, one model, one output — and run it 10 times.
• I see a 24-month window to build local-AI-native software for verticals that still run early-2000s tools.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

On this solo episode, I review five free, open source GitHub repos that help you build products, make money, or save time: Peter Yang's No AI Slop Skill, the CRM by TryComp AI, Video Use by browser use, SkillSpector by NVIDIA, and Phone Harness. For each repo I explain what it does, why it matters, how to install it, and the first small workflow to try. I close with a simple three-step method: install the repo, make one small workflow work, then decide to productize it or keep it as your own leverage.

Timestamps

00:00 – Intro

01:44 – Repo 1: No AI Slop

05:20 – Repo 2: Agentic-first CRM

10:52 – Repo 3: Video Use

15:15 – Repo 4: SkillSpector

18:49 – Repo 5: Phone Harness

22:25 – Closing Thoughts

Links to repos:

petergyang/no-ai-slop — https://github.com/petergyang/no-ai-slop

trycompai/crm — https://github.com/trycompai/crm

browser-use/video-use — https://github.com/browser-use/video-use

NVIDIA SkillSpector — https://github.com/NVIDIA/SkillSpector

phone-harness — https://github.com/ShawnPana/phone-harness

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

In this solo episode I explain a role that I call the marketing engineer. I believe this person becomes one of the most valuable hires in tech in the next 18 to 24 months. I define the job, I show the four eras of marketing that lead to it, and I give the tool stack that makes it work. I use a commercial HVAC software company as a worked example, and I list six systems that a marketing engineer builds. I close with four ways to earn money from this skill and a 30-day plan to learn it.

Timestamps:

00:00 – Intro

01:46 – The Evolution of Marketing

04:29 – What is a marketing engineer

07:19 – Build the Growth OS

10:18 – Marketing Engineer Tool stack

13:23 – Live Data Workflow

14:32 – Agent Job Description

16:56 – Example: vertical SaaS for HVAC contractors

18:27 – System 1: Customer Truth

20:20 – System 2 - 4: Founder content, Outbound signal and Creative Testing

23:31 – System 5: AI search visibility and the growth cockpit

24:19 – System 6: Eval Loop

25:06 – Ways to Monetize

29:41 – The 30-day plan

32:24 – Closing Thoughts

Key Points

• I expect the marketing engineer to command salaries from 250K to more than 1 million dollars.
• I build the growth repo first, because it holds the marketing memory of the whole company.
• I write a job spec for each agent, in the same way that I write a job description for a person.
• I measure qualified replies and pipeline, because business results show the true signal.
• I treat taste and judgment as the moat, because agents become a commodity.
• I recommend one working system over five half-built ones.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

Vinny is back on the pod, and he explains WebMCP. WebMCP puts MCP tools inside the browser UI, so any agent that you bring can read a page and act on it. Vinny demos an espresso gear store where his agent compares two machines, checks the counter width, matches accessories, adds an item to the cart, and applies a coupon. Together we map the agent-native options, from headless APIs to in-app agents, and we place WebMCP in the middle of that map. I close the episode with two cash-flowing business ideas that a semi-technical founder can start today.

Timestamps

00:00 – Intro

02:28 – WebMCP Clearly Explained

07:22 – Demo: Conditional Tools And Browser Session Login

09:11 – The Agent-Native Paradigm

12:12 – How Agents Interact with Apps

16:05 – Demo: Accessories, Cart, And Coupons

18:42 – Where WebMCP fits

21:03 – Startup Idea 1: WebMCP Conversion Agency

24:27 – Startup Idea 2: Agent Mystery Shopper

26:11 – Closing Thoughts

Key Points

• WebMCP makes a website agent-readable and agent-actionable through a short, clear list of tools.
• The browser session carries the login, so tools stay conditional and the setup stays simple.
• The live store demo compares machines, matches accessories, adds to the cart, and applies a coupon.
• WebMCP sits between headless APIs and in-app agents, and it keeps the visual UI that most people prefer.
• Vince says WebMCP launched in February as a joint Microsoft and Google experiment in Chrome.
• The first markets: complex commerce, SaaS admin consoles, read-only flows in regulated industries, and internal tools.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

FIND VINNY ON SOCIAL:

X/Twitter: https://x.com/hot_town

Youtube: https://www.youtube.com/channel/UCHP5Hdx0X-sM0uv2bl_OOqg

On this week’s SIP Live we dive deep on the X timeline to give each post a sip or a skip, and we answer questions from the chat. This episode covers AI-native operations, a baby food app that earns a million dollars a month, iMessage agents, human customer support as a brand advantage, small teams, reading habits, and sales advice for builders. Listeners get concrete first steps, business ideas they can copy today, and two clear opinions on each take.

Timestamps

00:00 – Intro

00:59 – Where an AI-Native Company Starts

04:48 – The $1M-a-Month Baby Food App

10:31 – An App Store for iMessage Agents

16:32 – Customer Support Is Eating Engineering

23:38 – RIP to the 2-Pizza Team

26:36 – Paul Graham on Reading as an Advantage

33:31 – The Worst Moves for 2026

37:01 – From Designer to Salesperson

Key Points

• Make the company legible first: keep meeting notes and SOPs in files that an LLM can read.
• Let the team explore AI tools freely, then ask one department, such as admin, to lead.
• A million-dollar-a-month app leaves room for niche copies that earn 10K to 30K a month.
• Build loops that turn daily support transcripts into prototypes and measurable goals.
• Human support works as a marketing advantage, and premium buyers pay for it.
• Sales stays part of the owner's job, so learn the frameworks and start cold calling.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

FIND JONATHAN ON SOCIAL

Unscheduled CEO Podcast: https://www.unscheduledceo.com/

X/Twitter: https://twitter.com/Jicecream

LinkedIn: https://www.linkedin.com/in/jonathan-courtney-4510644b/

I sit down with Billy Howell for an inside look at a real business that runs on Grok Bot agent teams. Billy tests hundreds of tools for The Rundown, and he uses Grok Bot to run his local newsletter, The Arlington Bagel, which goes to 6,000 readers every Thursday. He walks me through his chief of staff bot, his research and sales agents, and the routines that move work forward while he sleeps. We cover the token math, the four-week plan he follows, and the business models that fit this tool best: newsletters, directories, and Shopify stores. You leave with a setup you can copy for one project this month.

Timestamps

00:00 – Intro

02:22 – Grok Bot Overview

03:12 – What makes GrokBot special

06:09 – Pick one project

08:09 – Creating your initial agent team

11:19 – First week sprint with Grok Bot

13:06 – Automated Routines within Grok Bot

15:51 – Agent Org Structure

17:35 – Best Practices for building Agents

19:45 – Newsletter playbook

21:12 – Plugins

22:24 – Newsletter playbook pt 2

25:43 – Sales agent

29:06 – Iterating and Improving Agents

33:34 – Grok Bot business ideas

36:45 – Site stack builder

38:36 – Directories and landing pages

41:42 – One project for a month

42:50 – Closing Thoughts

Key Points

• Keep one project per Grok Bot account, so context stays clean and tokens stay available.
• Start with a chief of staff bot, and let it audit your business and name the first three agents.
• Follow the four weeks: build the team, execute, hire and fire, then automate.
• Ask each agent for a five-line brief: what shipped, what is stuck, what needs you.
• Move repeatable steps into scripts or make com, and keep agent tokens for high value work.
• Newsletters and directories give the easiest entry, and the two models feed each other.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

FIND BILLY ON SOCIAL:

X/Twitter: https://x.com/billyjhowell

Instagram: https://www.instagram.com/billyjhowell

Tiktok: https://www.tiktok.com/@saasfatigue

Youtube: https://www.youtube.com/@billyjhowell

Get Your Complete Financial OS at https://startup-ideas-pod.link/brex_SIP

I talk with Remy, known online as AI with Remy, about agent skills and how to share them across a whole team. Remy explains that a skill is an SOP for AI: a markdown file that teaches Claude your exact way to do a task. He then shows his system. He keeps his team skills in one GitHub repository, and he installs that repository as a plugin in Claude Code and Codex. The result is one source of truth, automatic updates for everyone, version control, and company ownership of the work. Listen to this episode if you run daily tasks with agents and you want your whole team to get the same quality of output.

Remy’s prompt + skill setup: https://startup-ideas-pod.link/Remy-skills

Timestamps

00:00 – Intro

02:33 – Skills as SOPs for AI

04:27 – How the agent uses skills

05:13 – AI is single player

06:21 – Example Skills: Notion, Brand Voice, Email

08:07 – How to share skills

11:40 – What is a Plugin?

14:58 – Anatomy of a Plugin

15:30 – Version Control and Rollback

16:58 – A Second Repo for Personal Skills

18:29 – The Day Claude Deleted 150 Skills

20:02 – Skills as Company Assets

21:00 – A Web App on Top of the Repo

24:46 – How Many Skills to Build

27:00 – The Self-Improvement Loop

28:15 – Skill Maxxing

30:40 – Closing Thoughts

Key Points

• A skill is a markdown SOP that teaches an agent your exact way to do a task.
• Most skills sit on one laptop, so a great process stays with one person.
• A GitHub repo plus a plugin gives the team one source of truth for every skill.
• Auto-update sends each skill edit to every teammate in Claude Code and Codex.
• A company-owned repo keeps the skills when a teammate moves on.
• Remy builds thin agents and thick skills, so any harness can run the same process.

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

FIND REMY ON SOCIAL

X: https://x.com/remy_gaskell

Youtube: https://www.youtube.com/@aiwithremy

AI with Remy: https://www.aiwithremy.com/

Get Claude Code: https://startup-ideas-pod.link/claude-greg

In this solo episode I lay out the exact system I use to turn Claude Code into what I call an AI employee. My premise is simple: give Claude the same things you would give a person joining your company (a workspace, memory, a brief, a clear ticket, eyes, review, a schedule, and permissions). I build the whole setup live around a real idea I found on ideabrowser.com , a missed-lead responder for med spas, and I share the specific prompts I use at each step. By the end, the product, the customer feedback, the docs, the demos, the reviews, and the recurring work all live inside one operating loop. I close with a seven-day plan you can run at your own pace. And thank you to Claude and Anthropic for supporting the podcast.

Setup Claude Code to be your 24/7 employee: https://startup-ideas-pod.link/Claude-code

Timestamps

00:00 – Intro: The AI Employee Map

05:47 – Step 1: Creating The Workspace In Claude Desktop

14:53 – Step 2: The Brief And Plan Mode

18:08 – Step 3: The Ticket And Defining Done

22:15 – Step 4: The Eyes And Desktop Preview

26:13 – Step 5: Review In Layers And The Diff View

29:34 – Step 6: The Schedule And Routines

34:32 – Step 7: Parallel Agents And Worktree Isolation

39:14 – Step 8: Permissions: Safe, Ask First, Human-Owned

41:15 – Step 9: Skills, Connectors, And Hooks

44:28 – The Seven-Day Plan

47:42 – Closing Thoughts

Key Points

• I treat Claude Code like a new hire: workspace, memory, brief, ticket, eyes, review, schedule, permissions.
• The repo brain teaches Claude how I work and what good looks like.
• Plan mode comes first: Claude reads the context, proposes an approach, and waits for my approval.
• One ticket at a time means one task, one finish line, and one reviewable change.
• The eyes matter: Claude opens the app in desktop preview, clicks the flow, checks the console, and reports what a buyer experiences.
• Routines and permissions turn a chat tool into a 24/7 operator with clear boundaries.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

I sit down with Allie K. Miller to talk about the shift from managing AI agents to enabling them. Allie runs a workforce of 34 AI agents led by an AI chief of staff named Simon, plus six directors named after Friends characters. She shares her three-word prompt, her daily AI diary, her AI watchdogs, and her rule to build the factory before the product. We then debate the future of software: why enterprises still want a vendor to call, and why consumer software now rewards taste and distribution. Listeners leave with one mindset shift and a first step they can finish in under three hours.

Get Your Complete Financial OS at https://startup-ideas-pod.link/brex_SIP

Timestamps

00:00 – Intro

02:29 – Become a Great Agent Manager

04:49 – The Three-Word Prompt

08:12 – The Pyramid of Proactivity

12:23 – Making the Company Queryable

19:14 – How to Design an AI Workforce

22:25 – AI as a Watchdog

24:56 – Startup Opportunities

26:29 – Build the Factory, Then the Product

30:09 – The SaaS Question

34:53 – Consumer Software as Art

37:12 – High-Value Bottlenecks

44:56 – Closing Thoughts

Key Points

• Allie sits three rungs above her 34 agents. She sets the infrastructure and waits for escalations.
• Her strongest prompt runs three words on top of full business context: do smart things.
• She holds the risk tier steady and expands only the breadth and scope of agent work.
• A dictated daily diary captures the context that lives outside meetings, email, and Slack.
• AI watchdogs remain wide open: duplicate work, calendar conflicts, and meeting disagreements.
• The bigger play is a software factory. Build the primitives once, then ship each product faster.
• I see opportunity on both sides of software: enterprises want a vendor to call, and consumers reward taste.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/

FIND ALLIE ON SOCIAL

X/Twitter: https://x.com/alliekmiller

Instagram: https://www.instagram.com/alliekmiller/

LinkedIn: https://www.linkedin.com/in/alliekmiller/

In this solo episode I break down Cloudflare's AI agent announcement in plain English and explain why I see it as the new business model for the internet. I walk through AI crawl control, pay per crawl, the monetization gateway, and the x402 payment rail, and I show how a single request turns into a transaction. From there I give three startup ideas built directly on top of this shift: a niche data refinery, agent readiness for businesses, and expert archives turned into agent tools. For each idea I lay out the wedge, the first customer, the first version, and how I would sell it. My core claim: the internet is moving from pages humans visit to resources agents use, and the builders who move now own the doors.

Timestamps

00:00 – Intro

01:01 – The Old Paradigm of the Internet

02:57 – The New Paradigm of the Internet

03:41 – What Cloudflare is actually doing

06:33 – An AI Index for all our customers

07:34 – The Agent Internet Stack

09:09 – Why Now Is the Best Time to Build

10:29 – Startup Idea 1: The Niche Data Refinery

17:10 – Startup Idea 2: Agent Readiness for Businesses

23:43 – Startup Idea 3: Expert Archives as Agent Tools

30:36 – The Filter for Finding Ideas

32:33 – Closing Thoughts

Key Points

• The human web monetized attention; the agent web monetizes useful resources, priced per request.
• Cloudflare's pay per crawl, monetization gateway, and x402 turn the HTTP 402 status code into a live checkout at the edge.
• Idea 1: refine one niche's messy data into clean fuel for agents, starting with 100 businesses in one city.
• Idea 2: sell agent readiness by showing a founder exactly what AI says about their company today.
• Idea 3: package an expert's archive into one job-specific agent tool the audience already wants.
• Every one of these works as a manual services business today and productizes as agent payments mature.

The #1 tool to find startup ideas/trends - https://www.ideabrowser.com

LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/

The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/

FIND ME ON SOCIAL

X/Twitter: https://twitter.com/gregisenberg

Instagram: https://instagram.com/gregisenberg/

LinkedIn: https://www.linkedin.com/in/gisenberg/