In pharmaceutical R&D, data quality issues do not always start in the data layer. A stale dataset, missing feature, or failed dashboard may trace back to degraded infrastructure, pipeline failures, upstream dependency issues, or changes in the underlying platform.

In this session, Naveen Kottala (Head of R&D Platform Engineering) and Ze Wang (Head of R&D Data Engineering) will share how they built observability across their R&D ecosystem in two layers: platform observability first, then data observability on top. You’ll learn how their teams created visibility across a complex environment spanning laboratory and clinical data flows, portfolio systems and cloud data platforms running on AWS and Snowflake.

They will cover how Sanofi correlates infrastructure and pipeline telemetry with data quality signals such as freshness, volume, schema drift, and distribution anomalies to speed up root cause analysis. They’ll also show how end-to-end lineage, data ownership workflows, and data monitors help teams trace issues from source systems and transformation layers to downstream datasets, scientific dashboards, and AI workflows.

You'll leave with a practical framework for building observability across both platform and data layers, enabling faster issue detection and triage while establishing a more reliable foundation for data analytics and AI.