Fix my web app deployment and stop the AI feature from crashing (AWS, Docker, CI/CD)

Hi, I have a small customer support web app. It uses Next.js, Node.js, and PostgreSQL, and it is hosted on one AWS server. Most of the code was written with AI tools like Cursor and Claude. It works, but it is not stable for real users. These are my problems: I deploy by logging into the server and restarting the app by hand. It breaks often. MyAI feature uses the OpenAI API. When OpenAI is slow or gives an error, the app hangs and sometimes crashes. MyAPI keys and .env file are saved in my GitHub repo. I have no alerts, so I only find out about downtime when a customer tells me. What I need: 1. Review my code and server, then send me a short plan before you start. 2. Move all secrets out of the repo and into AWS. Tell me which keys I need to change. 3. Add Docker for the frontend and backend. 4. Set up GitHub Actions so every push to main runs tests and deploys automatically. 5. Fix the AI feature with a timeout, retries, and a simple error message for users. 6. Add a health check and one alert that emails me when the server is down. 7. Write a short guide so I can deploy and roll back by myself. Not included: new features, UI changes, or moving to another cloud. Payment: Fixed price, $150. I will pay in two milestones: $50 after the plan and secrets cleanup, and $100 after everything works on my server. Timeline: 5 to 7 days In your proposal, please answer these two questions: 1. Have you set up GitHub Actions to deploy to AWS before? Tell me about one project. 2. How would you handle OpenAI rate limit errors in a Node.js app? Thank you.