Architected and built from scratch the core delivery coordination microservice for WhatsDish platform using Go and Hertz framework. Optimized PostGIS geospatial queries achieving sub-100ms response times for driver proximity searches, enabling real-time driver assignment based on location. Integrated Google Maps APIs for multi-waypoint route calculation with distance matrix optimization and polyline encoding for turn-by-turn navigation. Designed payout calculation engine accounting for base rates, distance-based fees, and time bonuses with dynamic rate adjustments. Implemented Protocol Buffers for type-safe API contracts with code generation. Built WebSocket infrastructure supporting 100+ concurrent driver connections for real-time location tracking. Deployed production infrastructure on AWS ECS Fargate using Terraform with auto-scaling, RDS PostgreSQL, and Grafana monitoring.

Real-time delivery coordination microservice built with Go, handling driver tracking, route optimization, and automated payouts.

WhatsDish Delivery Microservice

An event-driven image processing pipeline leveraging Spring Boot and AWS Lambda. The architecture decouples high-throughput ingestion from CPU-intensive processing using SNS/SQS buffering. It features a REST API for uploads, asynchronous Lambda functions for image transformations (resizing, watermarking), and uses S3 for storage and DynamoDB for metadata. The system includes end-to-end distributed tracing with OpenTelemetry and Grafana for observability, allowing visibility across microservices. Infrastructure is managed via Terraform and supports local development with LocalStack.

Event-driven image processing pipeline built with Spring Boot and AWS Lambda, featuring distributed tracing.

Projects tagged as "terraform"