Streamlining Insurance Intake with Amazon Nova Pro and Amazon Bedrock Agents by HCLTech
HCLTech partnered to build a GenAI-powered Intelligent Insurance Intake solution leveraging Amazon Textract, Amazon Bedrock Nova Pro Large Language Model, Amazon Bedrock Agents, AWS Lambda, and Amazon DynamoDB to automate processing of complex insurance forms. The solution handles diverse form layouts with a hybrid approach combining structural AI with contextual LLM understanding, achieving about 95% extraction accuracy and up to 20X process time reduction. Deployed for a leading Canadian insurance provider to automate workers' compensation form processing, reducing manual staff time from 60 minutes per form, lowering error rates, and improving customer satisfaction. Serverless architecture enables scaling and flexible workflow configuration with data securely stored and compliant with HIPAA and GDPR regulations.
- Organization
- Leading Canadian insurance provider
- Industry
- Insurance
- Location
- Canada
- Published
- May 2025
Reported outcomes
−95%
timeTime & speed
Strategic outcomes
Catalog median for time & speed deployments: −60% across 727 reported metrics. Compare benchmarks →
Primary read
Use case focus
Showing 2 of 2
- 1Document Processing Automation
- 2Workflow Automation
- Insurance companies face costly and inefficient manual processing of complex forms that contain structured and unstructured data with high error rates.
- Maintaining form-specific models and integrating diverse form revisions and layouts is challenging to scale.
- Operational delays from manual data entry impact customer experience and compliance risks.
- Implemented a serverless AWS architecture combining Amazon Textract for data extraction and Amazon Bedrock Nova Pro LLM to interpret and transform the data contextually.
- Amazon Bedrock Agents orchestrate multi-step workflows via conversational natural language interface to insurance professionals.
- AWS Lambda functions manage business logic, and Amazon DynamoDB stores processed data in a flexible, insurance-optimized schema.
- Security enforced through encryption (TLS, KMS), IAM access controls, CloudTrail audits, GuardDuty monitoring, and Amazon Bedrock Guardrails to ensure responsible AI usage.
- Achieved a 20X reduction in insurance form processing time and boosted extraction accuracy to 95%.
- Reduced operational costs and manual workload with minimal human intervention.
- Improved competitiveness by enabling faster and scalable processing across form types.
- Enhanced customer satisfaction and adoption by streamlining underwriting and claims workflows.
Architecture
Serverless architecture on AWS using Amazon Textract for document structure extraction, Amazon Bedrock Nova Pro LLM for contextual data interpretation, Amazon Bedrock Agents for workflow orchestration, AWS Lambda for business logic, and Amazon DynamoDB for data storage, with enterprise-grade security controls.
Implementation partners1
Sources & evidence1
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Was this useful?