I built a team of AI Agents that grow my business 24/7 (full demo)

April 14, 2025

AI Summary

5 min read

I Built a Team of AI Agents That Grow My Business 24/7

The gap between what's possible and what people are doing

Flo, CEO of Lindy.ai, argues there is currently a "huge arbitrage" between what AI agents can do and what most business owners are actually using them for. The companies exploiting this gap are "blowing up" — he points to a startup called The Arcades that reportedly hit $5 million in ARR in a few months with roughly eight people. The implication is clear: the frontier has moved, but most people are still "using the fax machine."

Start with the personal assistant stack

Flo recommends beginning with the most universal use cases: meeting recording, meeting scheduling, and meeting preparation. These are easy to set up, immediately useful, and build confidence. His own meeting recorder is a good example of how agents compound in value over time. It records meetings, summarizes them, and then looks up past Google Docs for the same person to append notes to the existing document. When his new head of marketing asked for context on past conversations with agencies, Flo could send a single Google Doc containing everything they'd ever discussed. "It just works," he says.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

Never miss an episode of The Startup Ideas Podcast

Get every new episode summarized in your inbox — free, ~5 minutes to read.

No spam. Unsubscribe anytime.

What you'll learn

1 (00:55) **Introduction & Episode Promise** - Flo returns to demo Lindy.ai, a no-code AI agent platform, and promises viewers they can automate half their business in half a day and create their first agent in 10 minutes.
2 (01:27) **The Simplest Use Case: Meeting Recording** - Flo starts with the most universal AI agent use case: recording and summarizing meetings.
3 (03:25) **Agents vs. Workflows: The Key Difference** - Flo explains that unlike Zapier, Lindy agents are AI-native, not just a sequence of isolated steps.
4 (04:41) **Live Demo: Building an Agent in Minutes** - Flo creates a simple agent from scratch to scrape YouTube comments, showing how little configuration is needed.
5 (06:31) **The "Exocortex": Using Agents as an External Brain** - Flo demonstrates how he uses his meeting recorder agent as a searchable memory for past conversations.
6 (08:08) **The Executive Assistant Stack & Agent Swarms** - Flo explains how he has replaced 70% of his executive assistant's work with a "chief of staff" agent.
7 (10:18) **Deep Dive: Agent Swarms (The "Agent Smith" Effect)** - Flo explains why agent swarms are necessary for reliability and speed at scale.

+ Full timestamped outline available in the app

Show Notes

Use agents https://www.lindy.ai/greg (thanks to Flo for hooking us up with 50% off and you can get started for free)

Join me as I chat with Flo Crivello, founder of Lindy AI, where he demonstrates how to build AI agents that can automate various business processes without coding. The conversation showcases practical applications including meeting recording, scheduling, recruitment, competitive analysis, and customer support. Crivello emphasizes that there's currently a significant gap between what's technically possible with AI agents and what most businesses are implementing.

Built your first AI Agent with Lindy: https://www.gregisenberg.com/ai-agents

Timestamps:
00:00 - Intro
02:08 - Demo 1: Meeting Notes
04:41 - Demo 2: YouTube comment scraping
06:49 - Meeting Notes Continued
08:27 - Demo 3: Outbound Phone Calling
10:16 - Demo 4: Meeting Prep
10:25 - Agent Swarms Explained
11:25 - Meeting Prep Continued
13:31 - Demo 5: Meeting Scheduler
15:04 - How to start using Lindy
15:52 - Live Building of a Recruitment Agent
17:57 - When should you loop in a human
18:37 - Building of a Recruitment Agent continued
20:34 - Demo 6: Sales Prospecting
22:44 - Why start using AI Agents
26:00 - Overview of template categories and use cases
28:52 - Demo 7: Elon Lindy
30:12 - Demo 8: Competitive Analysis
32:58 - Demo 9: CRM and Networking Manager
34:50 - Final Thoughts

Key Points:

• Lindy is a no-code platform for building AI agents that can automate business processes
• Users can create their first agent in 10 minutes and automate significant portions of their business
• The platform features "Agent Swarms" that can handle multiple tasks in parallel with greater reliability
• Lindy offers thousands of integrations with various platforms and services

1) Meeting Assistant Agents

The most universal use case for AI agents? Meeting management!

Lindy can:
• Record and transcribe your meetings
• Create searchable notes in Google Docs
• Organize notes by person (incredible for context)

Flo: "I just had to dig up Google Docs and send him everything we've ever talked about with this person."

2) The Agent vs Workflow Difference

Unlike Zapier which connects isolated steps, Lindy creates TRUE AGENTS that:

• Understand context between steps
• Can recover from mistakes
• Let you speak to them in natural language
• Don't need every step configured

The magic? You're basically telling an AI to execute a sequence of connected actions.

3) Agent Swarms = GAME CHANGER

This new feature lets you deploy multiple agents simultaneously!

• Send personalized outreach to hundreds of leads
• Research multiple people in parallel
• Execute tasks reliably without losing coherence

Think Agent Smith in The Matrix - your agent duplicates itself to ha