AntreAja
Nationwide digital queue management system for Telkomsel GraPARI outlets
Context
AntreAja is a nationwide digital queue management system operating under the DSC ecosystem, built to replace physical queue-taking at Telkomsel’s GraPARI outlets with remote, trackable virtual queues across 500+ outlets nationwide. Built as a Senior Full Stack Developer at PT. Dans Multi Pro, within Telkomsel’s ecosystem.
Technical challenge
Queue state needed to stay accurate and real-time across a nationwide network of outlets simultaneously, processing thousands of queue events a day — customers checking queue status remotely needed to see the same state staff saw on-site, with no lag that would send someone to an outlet just to find the queue had moved on.
Architecture & decisions
Golang microservices handle the full digital queue flow — creation, assignment, status updates — with Kafka providing event-driven communication between services so status changes propagate reliably under high volume. MySQL and Redis back persistent and fast-access queue state respectively. Building this as a nationwide, event-driven service from the start — rather than a per-outlet queue tool — was the deliberate scope decision, since Telkomsel’s outlet network made a fragmented per-location system a non-starter operationally.
Contribution
Maintained and extended the Golang-based microservices handling the end-to-end queue flow, working across both backend and frontend to ship features using Golang, REST APIs, MySQL, and Redis. Monitored and troubleshot production services using Splunk logs to investigate issues and keep daily operations reliable across 500+ outlets.
Result
Customers can locate the nearest outlet, book a queue number in advance, and monitor service availability remotely — removing the need to physically queue to know where they stood — while the platform reliably processes thousands of queue events every day across 500+ outlets nationwide.
Lesson
Real-time queue state is a distributed-systems problem disguised as a simple feature; putting Kafka between services early meant status updates stayed consistent even as event volume scaled nationwide.
Impact
- GraPARI outlets served
- 500+
- Queue events processed
- 1,000s/day