Skip to main content

WardMitra - Introduction

WardMitra is SPW Healthcare Innovations' citizen grievance, ward operations, and service-tracking platform. This documentation set is being shaped as the operating reference for the SPWHI team, starting with the target infrastructure architecture required to move from today's manually managed deployment model to a supportable AWS platform.

Live app: wardmitra.spwhin.com

Product context

WardMitra is expected to support:

  • municipal ward operations and issue tracking
  • citizen-facing workflows with traffic spikes during campaigns and ward rollouts
  • backend processing for complaint assignment and AI-assisted analysis
  • future scale in the range of 30,000 to 50,000 daily active users and 500 to 800 concurrent users

Current-state concerns

The current operating model is not strong enough for long-term scale. The main risks called out for this workstream are:

  • manual deployment steps
  • no consistent staging path before production changes
  • weak secrets and environment management
  • limited backup and recovery posture
  • limited monitoring and alerting
  • no clear isolation model for AI or asynchronous workloads

Target platform summary

The target direction for WardMitra is:

LayerTarget direction
FrontendReact/Vite static build served from S3 + CloudFront
Backend APINode.js/Express on Amazon EKS
DatabaseAmazon RDS for PostgreSQL 15 with PgBouncer
AI / async workloadsIsolated workers on EKS, with dedicated scaling controls
InfraAWS VPC + EKS + managed AWS services, provisioned with Terraform
DeliveryGitHub Actions with OIDC-based AWS authentication
EnvironmentsSeparate dev, staging, and prod paths

What to read first

Guidance for the current SPWHI team

This portal is being written for a delivery team that is still growing its DevOps depth. For that reason, the first deliverable intentionally favors:

  • simple repo boundaries
  • Terraform-only infrastructure management
  • GitHub Actions OIDC instead of long-lived AWS keys
  • managed AWS services where possible
  • progressive rollout from a basic dev setup to a full multi-environment platform