
Fix Critical AWS Website Bugs
- or -
Post a project like this- Posted:
- Proposals: 53
- Remote
- #4507815
- OPPORTUNITY
- Open for Proposals






Description
The components that must be repaired:
• Email verification (custom-built)
• Dialer system (Twilio)
• Booking system (custom-built)
• Payment system (Stripe)
• Live functionality / real-time chat
All code will stay where it is—nothing leaves AWS—so you’ll need solid AWS console familiarity alongside everyday full-stack skills. I can provide repository access, architecture notes, and direct contact with the previous developer if needed.
Acceptance is simple: each item above must perform reliably under normal user load, pass my own end-to-end test scripts, and remain stable for 48 hours in staging before we flip the live switch. If any hotfixes are required after launch, I expect a prompt turnaround.
Please outline your approach, relevant AWS + Twilio + Stripe experience, and confirm you can commit to the 14-day window.
ASHLEY G.
0% (0)New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-

Can you share the current AWS architecture (e.g., EC2, ECS, Lambda, Amplify, RDS, API Gateway, CloudFront) and confirm whether the failures are isolated to application logic, third-party integrations (Twilio/Stripe), or infrastructure issues such as IAM permissions, networking, environment variables, or WebSocket configuration?
-

Hi Ashley,
Before getting started, I'd like to clarify a few details:
1. Which AWS services are you using (EC2, ECS, Elastic Beanstalk, Lambda, EKS, Lightsail, etc.)?
2. What is your application stack (Node.js, .NET, PHP/Laravel, Python, React, Vue, etc.)?
3. Is the real-time chat built with WebSockets, Socket.IO, AWS API Gateway, Pusher, or another service?
4. Are these issues isolated to production, or do they also occur in staging?
5. Do you already have centralized logging enabled (CloudWatch, Sentry, Datadog, etc.)?
I look forward to working with you. -

One question: Are the five components failing with specific error codes you can share, or are they failing silently with no visible error, as that single distinction tells me immediately whether this is a configuration issue or a logic bug and cuts the triage time in half?