Agenda
- 5:30 - 6:30pm Networking
- 6:30 - 7:30pm Presentation
- 7:30 - 7:45pm Closing
Abstract
In today’s enterprises, AI initiatives often falter—not due to poor models—but because of brittle data foundations. This session explores how robust data engineering pipelines unlock scalable, trustworthy AI. Drawing on real-world transformations in manufacturing and finance, I’ll showcase patterns for integrating multi-terabyte transactional systems with Azure-hosted data lakes, and how we employed CI/CD-driven ETL to reduce nightly workloads from 7 hours to 2. We’ll discuss how microservice-based APIs, domain-driven design, and data validation frameworks directly impact ML feature quality and compliance. Crucially, we’ll delve into how incorporating data lineage, anonymization, and role-based masking upholds data privacy and fuels responsible AI outcomes.
This talk provides actionable guidance for leaders and engineers seeking to bridge the gap between operational systems and ML pipelines—delivering not just clever models, but resilient architectures that adapt to shifting data and regulations.
Bio
Bhaskar Bharat Sawant is a seasoned technology professional with over 15 years of combined academic and industry experience spanning artificial intelligence, machine learning, and cloud-native software engineering. His work integrates cutting-edge domains such as big data analytics, predictive modeling, IoT, and cybersecurity to deliver intelligent, scalable solutions across enterprise environments.
