Loading...
Menu
Loading...
The Real Estate WhatsApp GPT Bot is a chatbot solution that analyzes real estate listings and basic user inquiries via WhatsApp using GPT, generating automated responses. It engages with users in natural language and can evaluate listings and answer info-related questions.

Focus
Data · Product · Software
Output
Model / Dashboard / MVP
Stack
Python · OpenAI API · Automation · WhatsApp Business API
Click image to enlarge
Project Details
I've documented the problem, approach, experiments, and outcomes here in an organized manner.
When building this project, my goal was to create a GPT-powered bot capable of natural language interaction with users via WhatsApp. The bot can respond to requests about real estate listings—such as pricing, basic info, or summaries—by analyzing user messages and generating automated replies. Incoming WhatsApp messages are processed, analyzed, and sent to a GPT model; then the generated reply is formatted and sent back.
This setup allows processing listing details using a predefined prompt set and delivering contextual responses as part of conversational flows.
The primary goal of this project was to build a ChatGPT integration capable of natural language interaction via WhatsApp that can understand and respond to real estate listing inquiries. Initially, parsing the incoming WhatsApp API data correctly and accurately relaying user intents to GPT was challenging. Prompt engineering was necessary to ensure the bot produced correct responses given the linguistic variety in listing descriptions.
I iterated on normalizing message formats and handling longer text inputs by developing strategies for chunking and managing conversational state. This process gave me concrete experience with real-time NLP, API integrations, and GPT-driven response mechanisms.