HomeServicesResultsAboutProcessBlogBook a Free Auditarrow_forwardhello@joropi.com
Client Results

Real Systems. Real Outcomes.

Every project listed here is a system we built for a real client. We don't pad portfolios with concepts or mockups. These are the results our AI systems delivered.

AI Voice Receptionist — Commercial Cleaning Co.
$2,000 Contract Closed
AI Automation

AI Voice Receptionist — Commercial Cleaning Co.

A commercial cleaning company was losing after-hours enquiries to voicemail. We deployed an AI voice receptionist that answered calls 24/7, qualified callers based on job size and location, and booked site walkthroughs automatically. At 11:47pm on a Tuesday, the agent answered an inbound call from a facility manager looking for ongoing warehouse cleaning. It qualified the call, booked a 10am walkthrough, and sent a confirmation — without any human involvement. Three days later, the client signed a $2,000/month ongoing contract.

$2,000
Contract Value
11:47pm
Call Time (owner asleep)
3 Days
Time to Close
Voice AIAppointment Booking24/7 CoverageVapi
30-Second Lead Qualification Pipeline
< 30s Response Time
Lead Systems

30-Second Lead Qualification Pipeline

A service business was generating leads through paid ads but converting poorly. The diagnosis: average response time was over 2 hours. We built a full qualification pipeline: Lead fills form → Make.com captures in Airtable → AI scores as Qualified/Not Qualified → If qualified, voice agent calls within 30 seconds → call outcome logged → Gmail alert sent: 'New qualified lead ready to close.' Zero manual work. Every qualified lead contacted in under 30 seconds. Always.

< 30s
Lead Response Time
100%
Response Rate
0 hours
Manual Follow-Up
Workflow AutomationVoice AICRM IntegrationMake.comVapi
STO-Bot — AI Quotation System for Auto Repair
Manual Research Eliminated
Custom AI Tools

STO-Bot — AI Quotation System for Auto Repair

An automotive repair company in Russia spent several hours every day manually searching parts prices and market rates before quoting customers. Errors were common. Quotes were slow. Customers were left waiting. We built STO-Bot: a custom AI system where a repair job is entered into a dashboard, the AI searches everything needed automatically, and generates a full itemised report in seconds. The owner edits, finalises, and shares with the client — all from one screen. Clients can also make their own search requests and submit repair enquiries via Telegram. The system self-learns: every correction the owner makes trains the model to improve.

Seconds
Time to Quote (was: hours)
Self-Learning
Improves with every use
2 Channels
Telegram + Web
Custom AI ToolPythonNext.jsOpenAI APITelegram Bot
RAG Knowledge System — Internal Documentation Search
5 Years of Docs → Instant Answers
Custom AI Tools

RAG Knowledge System — Internal Documentation Search

A company with 5+ years of technical documentation needed a faster way for their team to find accurate answers. Staff were spending hours searching through folders for information they needed in real time. We built a RAG (Retrieval-Augmented Generation) knowledge system that connects to their Google Drive and internal databases, converts everything into searchable vector embeddings, and returns accurate answers — with citations to the exact source. The system updates automatically when new documents are added.

Seconds
Answer Time (was: hours)
100%
Source-Cited Answers
Auto-Updates
When new docs added
RAGLangChainOpenAI APIGoogle DrivePython

Ready to Be the Next Case Study?

We take on a limited number of new clients each month to maintain quality. If you're a service business losing leads to slow response times or manual processes, book your free audit and let's see if we're a fit.