A save fails, a payment gets stuck or a change breaks another feature. Tell us what happens and what you expected instead. We can inspect the code and propose a fix.
Find · Reproduce the issue
We start with the steps that trigger the error and the expected result. For intermittent failures, we check logs, data and the environment. If those don’t tell us enough, we add the diagnostics needed to investigate.
- Broken onboarding, billing and account workflows
- Data consistency and background processing failures
- Authorization defects and production configuration issues
Fix · Make the change
We agree on the work, and we deliver changes in your repository for you to review. The fix might be a condition, a query or a transaction. We explain what needs changing and why.
Re-Test · Verify the fix
We repeat the original failure and check the other workflows affected by the change. Where useful, we add a regression test to catch the issue in future releases. The handoff includes deployment instructions.
Ship · Release and support existing users
We first consider the impact on customers and how to contain the problem. We agree on where to test and how to roll back. Restoring service may come before a longer-term fix.
Before we start
Can you help with a single bug?
Yes. Describe the behaviour, the steps to reproduce it and the expected result. Before agreeing on a fix, we assess whether a separate investigation is needed.
Do you work on apps that already have users?
Yes. We agree on a test environment, access and release procedure. If customers are affected, we first consider how to contain the issue and restore service.
Will the whole application need rewriting?
Not necessarily. We start with the failure and the affected components. We propose refactoring when it is needed to address the issue or verify the fix.
Updated