Indegene Builds Generative AI Data Quality Agent on AWS

Life sciences organizations face challenges maintaining consistent data quality across complex platforms and regulated domains. Indegene Limited built a production-grade Data Quality Agent within its NEXT Unified Data Platform hosted on AWS to address limitations in conventional static validation logic. The solution leverages Amazon Bedrock, Strands Agents, and Amazon Bedrock AgentCore, allowing the agent to suggest context-aware business and technical validation rules based on data profiling, anomaly detection, and two retrieval-augmented generation knowledge bases: an internal enterprise repository and an external reference source. Generated rules are reviewed by data stewards before producing executable SQL and Python code, storing validated logic in a central repository for continuous monitoring and governance.

Key takeaways

• Indegene deployed an LLM-driven Data Quality Agent inside its NEXT Unified Data Platform on AWS, utilizing Amazon Bedrock, Bedrock AgentCore, and Strands Agents.
• The agent utilizes two RAG knowledge bases—internal enterprise documentation and external reference standards—to generate context-aware business validation rules beyond simple null or type checks.
• A human-in-the-loop workflow guides users through source profiling, anomaly detection, rule recommendation, review, and automated generation of validated SQL and Python code.