Project
DocEngine
A distributed document-processing platform that turns header-based CSV uploads into PDFs through asynchronous jobs, with tenant isolation, object storage, and live operational monitoring.
Project Overview
DocEngine accepts authenticated CSV uploads, stores source files in object storage, records tenant and job metadata in PostgreSQL, and hands work to RabbitMQ for asynchronous CSV-to-PDF processing. Clients can track job status and receive short-lived download links when results are ready. The deployed system includes a React frontend, Spring Boot API and worker, Supabase PostgreSQL, Amazon S3, and Prometheus/Grafana monitoring.
Key Features
- •Header-based CSV uploads with flexible columns
- •Asynchronous job processing through RabbitMQ
- •Tenant API-key authentication and isolated job reads
- •Idempotent submissions and atomic worker job claiming
- •Lease and heartbeat recovery for processing jobs
- •PDF result storage with short-lived download URLs
- •Prometheus metrics and Grafana dashboards
Project Details
Date
January 2026
Deployment
Production deployment- Frontend: Vercel — docengine1.vercel.app
- Spring Boot API and CSV-to-PDF worker: AWS EC2 with Docker Compose
- RabbitMQ, Prometheus, and Grafana: AWS EC2 with Docker Compose
- PostgreSQL: Supabase
- Uploaded CSV files and generated PDFs: Amazon S3