Shrikant Sonparote

Lead Engineer, Information Security

Lowe's Companies, Inc.

Roadblock Title:

How to Use Shadow/Replay Testing to Catch Production Bugs Before They Ship

Time:

Wednesday - 2:00 PM (Tower A)

Abstract:

Fintech systems often face a reliability gap: staging is necessary, but it rarely captures the production edge cases that cause late regressions, on-call fatigue, and customer-impacting incidents.

Drawing on 19 years of experience across enterprise, banking, and high-scale reliability environments, this Roadblock Talk presents a practical three-layer validation model for high-stakes systems: CI/CD-driven ephemeral environments, legacy-compatible staging, and a governed pre-prod shadow/replay layer.

The session shares an implementation blueprint for safely converting production-shaped behavior into release evidence through traffic capture, security masking, replay orchestration, side-effect isolation, Playwright-based regression flows, business-invariant comparators, and CI/CD scorecards.

Attendees will leave with patterns for reducing release risk in regulated systems while protecting customers, money movement, and sensitive data.

Bio:

Shrikant Sonparote is a Strategic Technology Leader, a Google Certified Professional Cloud Architect, and a BTA Certified Blockchain Solution Architect with over 19 years of experience delivering high-scale, resilient infrastructure for Fortune 50 and global financial giants. A veteran of the banking and FinTech trenches, he led JPMorgan’s blockchain-based trade matching initiatives and architected the mission-critical cashback framework for Fidelity’s Bloom app, achieving a 90% improvement in scalability. Currently, at Lowe’s, he serves as the lead technical authority for identity security and governance. In this role, he engineered a resilient authentication framework, reducing latency by 40% while maintaining 99% uptime across 1,700 store locations. He is a Senior Member of the IEEE and is currently dedicated to establishing the security and reliability guardrails for the next generation of Agentic AI microservices and Non-Human Identity (NHI) management.