No more typing reviews! Try our Samantha, our new voice AI agent.

AI-Powered Insurance Data Warehouse & Analytics Platform

HG
AI ENGINEER at a insurance company with 501-1,000 employees
200 people affected
1 people managed
6 month project

Project Description

Developed an enterprise-grade AI and cloud analytics platform for the insurance industry. The project automates insurance data ingestion, ETL processing, document intelligence, and analytics using AWS Glue, Amazon S3, Python, and Generative AI. It processes data from multiple insurance providers, converts it into optimized Parquet datasets, and supports business intelligence dashboards for operational monitoring. The platform significantly reduced manual effort, improved data quality, and accelerated reporting for business teams.

Lessons Learned

If I were to do this project again, I would adopt a lakehouse architecture from the beginning, implement Infrastructure as Code (Terraform), automate CI/CD pipelines, and use Amazon Redshift Serverless instead of provisioning clusters. I would also incorporate comprehensive data quality checks, monitoring, and AI-powered anomaly detection earlier in the development lifecycle to improve scalability, reliability, and operational efficiency.

Highlights

Ahead of schedule
Under budget
Support from colleagues

Difficulties

Management had to be convinced
Steep learning curve
Hard to meet schedule

Products Used

Technical Skills Used

  • Python AWS Data Engineering ETL PySpark Amazon S3 AWS Glue Amazon Athena Amazon Redshift AWS Lambda SQL MongoDB Generative AI LangChain REST API Data Warehousing Power BI Git
  • Ahmedabad (IN)23.025872.5873