Multi-Cloud Data Platform

Cloud-agnostic data platform architecture to onboard large ISPs on GCP while maintaining AWS parity.

## Context We needed to onboard large ISP customers on GCP while keeping the existing AWS platform reliable and consistent. The platform had to be cloud-agnostic without breaking existing pipelines. ## Constraints - Zero downtime migration for critical data flows. - Different cloud primitives and cost models. - High throughput ingestion and strict data quality expectations. ## What I built - Multi-cloud architecture and deployment strategy for the data platform. - Migration paths for streaming and batch workloads. - Governance and reliability guardrails to keep parity across clouds. ## Impact - Enabled onboarding for ISPs with 15M customers. - Reduced infrastructure cost by 30%. - Cut data quality issues by 50%. ## Stack GCP and AWS, streaming pipelines, and platform-level abstractions for portability.

Mission Brief

The Problem

New ISP customers required GCP, but the platform was AWS-native.

The Solution

A multi-cloud platform that preserved AWS parity while onboarding ISPs on GCP at scale.