endorsedCompleted

Machine-Learning Matching for Real Estate Sales Automation Platform

Services Covered:

AI and Machine Learning • Website Development • Software Maintenance and Support

Industry

Real Estate

Duration

3 months

Budget

50K - 199K

Start Date

1 July 2017

Team Size

6-10 Employees

Engagement Model

Offshore

Project Capability Score (PCS)

68/100

Good

Project Capability Score (PCS) estimates how capable Kindgeek is of successfully delivering a project like this, based on its past experience, track record, and ability to handle similar work. Learn more about PCS

Client

Confidential

San Francisco, United States of America

Project Summary

KindGeek developed the machine-learning matching mechanism for a Python-based sales automation platform for real estate developers, analyzing and categorizing property and user datasets and adding in-app analytics. They delivered the platform's defining ML feature and continue to provide ongoing support and development, contributing materially to company revenue.

Key Challenges

Improve machine learning to predict properties that would most interest clients.
Analyze and categorize datasets of properties and people to enable accurate matching.

Project Deliverables

AngularJS web application
Matching mechanism within the Python-based machine learning engine
Dataset analysis and categorization components
In-app analytics instrumentation
Ongoing developer support

Project Solution

Implemented the matching mechanism within a Python-based machine learning engine and developed the AngularJS web application front end. Applied a distant supervision approach to analyze and categorize datasets of properties and people, set up in-app analytics to measure component value, and provided ongoing support and additional developer resources for future releases.

Project Outcome

Machine learning functionality became the platform's main feature.
Company profits improved significantly and the feature accounted for the largest part of the revenue stream.

Platforms

Tech Stack

AngularJS
Python

Client Endorsement

Overall Review Rating

Timeliness

Cost Rating

Willing to Refer

Quality of Deliverables

“They built out a platform’s defining feature and are responsible for measurably increasing company revenue. KindGeek consists of talented individuals that are committed to delivering high-quality products. Executive level-engagement ensures frequent communication and a collaborative experience.”
Anonymous
CEO
Date: 07/06/2018