Scaling Dark Store Expansion and Quick-Commerce Fulfillment
How an automated expansion platform cut store setup time from 160 to 85 days across a 500-store national rollout.
01 The Business Situation
During India’s quick-commerce hyper-growth surge, Blinkit was scaling from hundreds of dark stores to over 500 fulfillment hubs nationwide. Rapid rollout velocity was the single biggest commercial advantage.
02 What Was Broken
Store expansion was manual: site procurement, equipment provisioning, catalog seeding, and network setup took up to 160 days per store.
Backend Golang microservices lacked uniform analytics, graceful shutdowns, and unified linters, causing telemetry drift across teams.
Mobile app releases were fragmented across disparate Android and iOS teams, slowing weekly release cycles.
Central data reporting suffered from ETL lag across dark store inventory and order allocation systems.
03 What I Decided & Why
Architected the Locus dark store expansion platform
Automated site onboarding, network provisioning, and dark store readiness checklists into a unified operational workflow, cutting setup time by 47% (160 to 85 days).
Built an enterprise-wide Golang service kit
Standardized graceful shutdowns, Datadog tracing, structured logging, and reusable HTTP/gRPC middleware, cutting new microservice launch time by 40%.
Spearheaded Kotlin Multiplatform (KMP) + Compose migration
Consolidated domain, network, and storage code into a single shared repository, achieving 80%+ code reuse across Android and iOS and cutting release cycles by 70%.
Built automated Apache Airflow data pipelines and Pinot virtual tables
Replaced manual reports with real-time operational dashboards for store ops, franchise onboarding, and on-call alerting.
04 Measurable Results
05 Lessons for Startup Founders
Do not maintain parallel mobile codebases if domain logic is identical. Kotlin Multiplatform or cross-platform code sharing saves months of executive synchronization.
Invest early in a shared service kit. If every engineer invents their own HTTP logger, metrics sink, and error handler, you will waste 30% of engineering time on plumbing.
Treat store and fleet onboarding as an automated software workflow rather than an ad-hoc operational spreadsheet.
06 Technologies Used
WORK WITH UMESH GUPTA
Ready to scale your architecture
without expensive trial and error?
Book a 20-minute call to evaluate your engineering bottlenecks. Response: Reply the same business day (IST).