Cloud-Native Data Pipeline and ML Optimization for Real Estate PM
Services Covered:

Real Estate
10 months
50K - 199K
1 September 2023
1-5
Onshore
Project Capability Score (PCS)
Project Capability Score (PCS) estimates how capable Valor Software 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
Skyland Transportation
Project Summary
Key Challenges
- Inefficient data pipeline and warehouse causing slow reporting during peak loads
- AWS Lambda functions and Redshift were not scaling and increased operational costs
- Generic ML model produced inconsistent predictions
- Global teams faced communication challenges due to time zone differences
Project Deliverables
- Cloud-native data pipeline on GCP
- Real-time ETL using GCP Dataflow
- BigQuery as primary data warehouse
- Custom ML pipeline using TensorFlow on GCP AI Platform
Project Solution
Project Outcome
- More accurate, tailored recommendations from the custom TensorFlow ML pipeline
- Improved customer engagement and conversion rates
- Reduced operational costs and improved analytics speed and accuracy
- Deliverables provided on time with strong communication
Platforms
Cloud
Tech Stack
Apache Beam
TensorFlow
Amazon RedShift
AWS SageMakerGoogle Cloud Platform (GCP)
Google Big Query
Client Endorsement
Overall Review Rating
Timeliness
Cost Rating
Willing to Refer
Quality of Deliverables
The new machine learning model from Valor Software gives more accurate recommendations, and our customer engagement and conversion rates have gone up. Even with the time zone differences, they were punctual, fast to respond, and kept in close touch over Slack and Jira. What they delivered was tailored to us.
Note: This endorsement is based on publicly available client feedback from external review sources.

