Senior full-stack engineer: production hardening and migration off Replit

Take a live React + Node SaaS from Replit to a production-grade hosting setup We run a live web product for TV and digital creators (React front end, Node/TypeScript API, background workers, Postgres on Supabase, Redis on Upstash, Stripe billing, several third-party AI and data APIs). It currently builds and deploys from Replit with two deployments (web app and worker). We need an experienced engineer to make it production-ready and move hosting off Replit without downtime or data loss. Scope: Audit the current setup: build, deployments, environment variables and secrets, Redis and Postgres connections, background job workers, cron jobs, file storage, logging and alerting. Recommend and set up the target hosting (we are open to Fly.io, Render, Railway, AWS or similar; argue for one) with separate staging and production environments, infrastructure as code where practical, and a documented rollback path. CI/CD from GitHub: tests and type checks gate deploys; preview or staging deploys per pull request; production deploy on merge to main with a manual approval step. Secrets management (no secrets in the repo or build logs), environment parity between staging and production, and a written runbook for rotating keys. Observability: structured logs, error tracking, uptime checks, basic dashboards and alerts for the API, workers and queues. Reliability basics: health checks, graceful shutdown, worker restart behaviour, queue backpressure, database connection pooling, backups and a tested restore. Performance pass on cold start, bundle size and the slowest API routes; a load test against staging with a short written report. Cutover plan: DNS, custom domain, SSL, zero-downtime switch, and a post-cutover checklist. Decommission the Replit deployments only after sign-off. Documentation a non-engineer founder can follow: how to deploy, roll back, read logs, and where everything lives. What we are looking for: 5+ years shipping Node/TypeScript and React apps to production; you have personally run a migration like this. Strong with Postgres (Supabase a plus), Redis-backed job queues, Docker, GitHub Actions, and at least one of the hosts above. You communicate in plain language, write things down, and leave the system cleaner than you found it. How to apply: send two examples of production systems you have set up or migrated (what the stack was, what went wrong, what you changed), your hourly rate or a fixed quote for the audit phase, and your availability over the next 3 weeks. We will start with a paid audit (about one week) and a written plan, then contract the migration.