LLaMA2 powered LeasingAI Co-Pilot for 5X inbound leads conversion - ZebraAI

LLaMA2 powered LeasingAI Co-Pilot for 5X inbound leads conversion

Business Problem

  • Real Estate sales have been run the traditional way for decades, with long sales cycles, low conversion ratios, low responses & even lower personalization to the inbound sales leads. Uniti AI, an NY based SaaS real estate solutions provider, had a vision to leverage Inbound enquiries for properties to make the highest revenue point for their clients. Traditional Real Estate Businesses faced challenges in efficiently managing and responding to a high volume of inquiries and traditional automation led to non-personalized response generation, resulting lead drops. Human-like responses are crucial for maintaining a personalized and professional interaction with clients, but manual responses are time-consuming and prone to delays.
  • Uniti AI wanted to revolutionize this paradigm within the real estate world, building a hyper-personalized, human-like, relevant property management agent, to help their clients convert inbound leads faster, with high margins.

About ZebraAI

ZebraAI is building AI leasing representatives to help commercial real estate owners more efficiently and responsively manage leasing.


With the advent of Generative AI & the capabilities to generate content on the fly, customized to the prompts & data we can generate, GoML’s GenAI consulting team worked with the customer to build a product vision, which allows their clients to build hyper-personalized communication for end customers, leveraging user persona real time, resulting 10X increased conversion. The solution followed a 3-step approach. 

LLaMA2 powered LeasingAI Co-Pilot for 5X inbound leads conversion - ZebraAI
Click to View in Full Size


Inquiry Capture and Activation: Users submit inquiries through a web form or send’s an email; these are routed to the client’s inbox and trigger an AWS-powered workflow via Microsoft Power Automate. 

Data Management and AI Integration: Amazon RDS records the inquiry, and AWS Lambda, in conjunction with AWS Comprehend, processes the data, leveraging Claude’s NLP capabilities to craft human-like email responses. 

Automated Response and Human Oversight: The system sends AI-generated responses to the user, with a built-in option for manual review or editing before dispatch for personalized communication. 

Efficient Scheduling and Follow-up: Integration with scheduling tools automates appointment booking for property tours, streamlining the user experience and enhancing lead conversion. 

Continuous Monitoring and Version Control: AWS CloudWatch, along with CloudTrail, oversees the operation for performance and security, while Git and Docker maintain code integrity and facilitate deployment. 



Reduced response time
to inbound queries


Increased conversion of inbound queries, with increased relevance of property suggestions


Overall increased efficiency of the sales engine, with the auto co-pilot

Technology Stack​