SMILE Platform (UNDP for Kemenkes)
National immunization logistics system in production across all of Indonesia — streaming CDC + batch analytics platform.
Architecture
Real-time CDC streaming from the OLTP database into ClickHouse, then batch pipelines build the analytics gold layer.
Orchestration
Infrastructure
Monitoring
Overview
SMILE is a national immunization logistics system built for the Indonesian Ministry of Health (Kemenkes) with UNDP, running in production across every health facility in the country — from Sabang to Merauke.
On the OLTP side I built the streaming pipeline: change data capture from AWS RDS MySQL via a Debezium connector into Kafka, streaming ETL with RisingWave into Amazon S3, and finally into ClickHouse for analytics-ready storage.
On the ClickHouse side I built the batch pipelines that produce the gold layer, with analytics modeled in dbt and orchestrated by Jenkins and Dagster. The entire environment runs on Kubernetes, monitored with Grafana — and I also help maintain the cluster.
Highlights
- Real-time CDC pipeline: AWS RDS MySQL → Debezium → Kafka → RisingWave → S3 → ClickHouse
- Batch gold-layer pipelines and dbt analytics on ClickHouse
- Orchestrated with Jenkins + Dagster, fully containerized on Kubernetes
- Production-grade, national scale — every health facility in Indonesia
Role
Data Engineer — streaming & batch pipelines, analytics, cluster maintenance
Context
UNDP for Kemenkes (Indonesian Ministry of Health) — in production
Tech stack
More projects
View all →LidValid — Unified Data Validation Platform
Full-stack data validation platform with Tiered Validation — cheap aggregate checks across every table, then precise row-level checks only where they fail.
Dataklin — AI-Powered Data Quality & Entity Resolution
Data quality and entity-resolution platform that profiles, validates, deduplicates, and standardizes datasets before handoff — with LLM-generated rules.
Real-Time Fraud Detection
End-to-end real-time fraud detection pipeline: Flask UI → RabbitMQ → Kafka → Flink → ClickHouse → Grafana.