Case study 06 / AI Engineering · Automation
From a voice note to the right property.
A Telegram assistant that understands agent requirements and searches connected property data through an n8n workflow.
AI Real Estate Assistant
Workflow design, AI integration and data retrieval
01 / The challenge
Start with the problem.
Matching a client's requirements meant checking property listings one by one. Finding relevant records in a large owner dataset added another layer of repetitive searching.
02 / The approach
Connect the moving parts.
- 01
Built text and voice input through Telegram, including speech-to-text transcription.
- 02
Connected intent understanding to tools that search and filter business property data.
- 03
Combined property recommendations with permitted owner-record retrieval in one conversational workflow.
03 / Engineering decisions
The thinking behind the work.
Meet agents where they work
Used Telegram so a typed request or voice message could start the workflow.
Language into actions
Translated requirements such as location, bedrooms and budget into relevant data lookups.
Connected business data
Grounded retrieval in the connected inventory and permitted records available to the business.
04 / The outcome
What changed.
Brought property matching and record retrieval into an operational assistant, giving agents a simpler path from a client request to actionable information.
Shared on LinkedIn