Summary
Ferrumengine is a resilient application engine concept for dependable services, workflows and infrastructure orchestration. It bundles the platform primitives YoungShark uses on every engagement — service scaffolding, background jobs, observability, secrets, drift detection — so teams start a new product from a known-good baseline instead of a blank repository.
The problem
Every new backend starts from the same blank canvas: logging here, tracing there, an auth middleware, a job queue, a migration runner, a health endpoint, a deployment manifest. Engineering teams burn the first two weeks re-deriving the same primitives, and the result is never quite the same twice — which makes onboarding, audits and incident response harder than they need to be.
Our approach
Ferrumengine is an opinionated application engine: it provides the scaffolding (service skeleton, configuration, secrets), the runtime (background jobs, queues, schedulers), the operational layer (structured logging, OpenTelemetry tracing, health checks, drift detection) and the deployment surface (container build, infrastructure-as-code, CI/CD pipeline). Teams plug domain logic into well-defined extension points instead of wiring infrastructure from scratch.
Where it stands
- Service skeleton with built-in structured logging, tracing and health checks
- Background job runtime with at-least-once delivery and idempotency keys
- Drift detection across infrastructure and application configuration
- Reference deployment templates for AWS (Terraform) — work in progress
Stack
- Go 1.22
- TypeScript
- PostgreSQL
- Redis
- OpenTelemetry
- Terraform
- AWS
- Docker