I Let AI Cold-Call 100 Plumbers (Genspark)

Siraj Raval · Advanced ·🤖 AI Agents & Automation ·1h ago
Try Genspark (the AI agent platform from this video): https://bit.ly/4dmjMSl I let an AI cold-call 100 plumbers to sell them a 24/7 AI receptionist. My Stripe started at $0. I built this whole thing using Genspark by orchestrating 6 different AI agents from a single prompt: a research agent, an AI developer, a Stripe agent, a voice script agent, a call-for-me agent, and an AI inbox agent. The whole stack built itself, ran 100 live cold calls, and converted 4 plumbers to paid sign-ups in 1 hour. Here is exactly how it worked. ABOUT GENSPARK New registered users can get free credits. What's wild, Genspark is offering unlimited use of AI chat and AI image for all paid users in 2026. Unlimited subject to abuse guardrails. Try it: https://bit.ly/4dmjMSl TIMESTAMPS 00:00 The $0 Stripe Challenge 01:31 Why Plumbers (78% Go to Voicemail) 02:21 The 6-Agent Architecture from One Prompt 04:06 Running Genspark Live 04:53 Watching It Build the Landing Page (Pipedesk AI) 05:42 Lead Scraping Output 06:07 The Mission Control Dashboard 06:12 Live AI Cold Calls 06:30 First Conversions Land 06:44 4 Plumbers Sign Up in Stripe 06:54 Scaling This to Lawyers and Surgeons 07:13 What Else Genspark Can Do THE PROMPT I USED (FREE) Get the exact prompt that built this 6-agent system on GitHub: https://github.com/llSourcell/genspark-cold-call-prompt Note: you'll need a Genspark account to run it (new users get free credits with my link above). GENSPARK FEATURES USED IN THIS VIDEO 00:24 to 01:00 Multi-model access (GPT-5.5, Claude Opus 4.7, Gemini 3.1 Pro) 01:00 to 01:25 Genspark Super Agent template library (AI Developer, Slides, Sheets, Call-for-Me) 04:43 to 05:30 AI Developer Agent generates the full landing page 05:01 to 05:30 Stripe integration auto-wired into the landing page 05:30 to 05:42 SMS automation plus Calendly booking flow 05:42 to 06:02 Mission Control Dashboard generated by Genspark 06:07 to 06:30 Call-for-Me Agent running 100 calls in parallel 07:13 to 07:32 Slides
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

Related AI Lessons

OpenClaw Hit 250K Stars Faster Than React. I Spent 24 Hours Trying to Like It.
Learn from a developer's 24-hour experiment with OpenClaw, the terminal agent that surpassed React in GitHub stars, and discover its strengths and weaknesses
Dev.to AI
AiFinPay: Autonomous Payments for ruvnet/ruflo
Unlock seamless payments for AI agents with AiFinPay's one-line payment SDK, enhancing autonomous workflows and conversational AI systems
Dev.to AI
AiFinPay: Autonomous Payments for cirosantilli/china-dictatorship
Learn how AiFinPay's autonomous payment solutions can facilitate seamless transactions for sensitive platforms like cirosantilli/china-dictatorship
Dev.to AI
AI Copilot vs AI Agent Architecture - What's Actually Different (And Why It Matters)
Learn the key differences between AI Copilot and AI Agent architectures and why they matter for your AI projects
Dev.to · Pallavi Sharma

Chapters (20)

The $0 Stripe Challenge
1:31 Why Plumbers (78% Go to Voicemail)
2:21 The 6-Agent Architecture from One Prompt
4:06 Running Genspark Live
4:53 Watching It Build the Landing Page (Pipedesk AI)
5:42 Lead Scraping Output
6:07 The Mission Control Dashboard
6:12 Live AI Cold Calls
6:30 First Conversions Land
6:44 4 Plumbers Sign Up in Stripe
6:54 Scaling This to Lawyers and Surgeons
7:13 What Else Genspark Can Do
0:24 to 01:00 Multi-model access (GPT-5.5, Claude Opus 4.7, Gemini 3.1 Pro)
1:00 to 01:25 Genspark Super Agent template library (AI Developer, Slides, Sheets, C
4:43 to 05:30 AI Developer Agent generates the full landing page
5:01 to 05:30 Stripe integration auto-wired into the landing page
5:30 to 05:42 SMS automation plus Calendly booking flow
5:42 to 06:02 Mission Control Dashboard generated by Genspark
6:07 to 06:30 Call-for-Me Agent running 100 calls in parallel
7:13 to 07:32 Slides
Up next
AI that runs your repo in GitHub : New from Microsoft Research #ai #agenticai #github #workflow
Microsoft Research
Watch →