Inspiration

Blue-collar businesses lose thousands of dollars every month, not because demand is low, but because they can’t respond fast enough.

Customers expect instant replies. When someone has a leaking pipe or broken HVAC system, they don’t wait, they call the next provider.

We were inspired by a simple insight: Demand is there. Revenue is lost. The fastest responder wins.

We wanted to build an Agentic AI system that works 24/7 to capture, qualify, and convert leads into real jobs, automatically.

What it does

Our system acts as an autonomous AI agent for blue-collar businesses.

It:

  • Captures inbound leads via chat bot

  • Books appointments automatically

  • Logs structured lead data for analytics

  • Escalates urgent jobs instantly

Instead of missing after-hours calls or manually chasing leads, businesses get a fully automated lead-to-job conversion engine running 24/7.

How we built it

We built an agentic workflow using:

  • LLM-powered AI agent for natural conversation & qualification

  • Structured prompt engineering to control decision-making

  • Tool integrations (Pinecone for vector storage)

    • Postgres database logging system to track conversations and conversions.

The system operates in a loop:

  • Capture inquiry

  • Extract structured information

  • Evaluate lead quality

  • Take action (book / escalate / discard)

This transforms passive AI chat into an autonomous decision-making agent.

Challenges we ran into

  • Designing structured prompts that prevent hallucination while still sounding natural

  • Balancing autonomy vs. human override for urgent jobs

  • Handling edge cases (incomplete information, unclear service type)

  • Ensuring qualification logic was consistent across different industries

We had to iterate heavily on workflow design to make the AI act reliably like a real dispatcher.

Accomplishments that we're proud of

  • Everyone can create an agent for their business with just two clicks

  • Created industry-specific flows (plumbing, HVAC, electrical, etc.)

  • Abstracted technical details from everyday users

  • Most importantly, we built something that directly increases revenue for small businesses.

What we learned

  • Small businesses need simplicity, not complexity

  • Speed is everything in local service markets

  • Automation must be trustworthy to replace human workflows

  • Agentic AI is far more powerful when given clear decision boundaries

We also learned that the biggest opportunity in AI isn’t replacing jobs, it’s capturing missed revenue.

What's next for

  • Integrate voice AI for live phone call handling

  • Connect to CRMs and dispatch systems

  • Add dynamic pricing & quote estimation

  • Launch pilot programs with local service businesses

Our vision is to become the 24/7 AI workforce layer for service-based industries.

Share this project:

Updates