When the usual fixes stop working

AI coding tools are remarkably good at producing code, and remarkably willing to keep suggesting fixes. Sometimes those fixes move a problem around instead of solving it — especially when the real cause is outside the code.

Many of the hardest problems live in the spaces between things: between an application and its hosting, a domain and its DNS records, a database and its connection limits, a cache and the data behind it. Diagnosing them takes someone who has worked across all of those layers.

This is experienced technical investigation — not a magic repair service. We find the cause, explain it, and agree the best way to resolve it.

Typical problems

Problems we're well placed to investigate

Deployment and hosting

It works on your computer but fails once deployed
Poor or fragile deployment setup
Hosting configuration problems
SSL certificate, DNS or networking faults
Difficult migrations between hosts or platforms

Application behaviour

Confusing or recurring errors in an AI-generated application
Problems introduced by automated code changes
Authentication and sign-in issues
Broken API integrations
Database errors, slow queries or data problems

Performance and cost

Slow pages
Incorrect or missing caching
Unexpected server resource consumption or cloud bills
Infrastructure that has become more complicated than it needs to be

Two ways to resolve it — both valid

We'll recommend whichever is fastest and safest for your situation, and you decide.

Option one

We fix it directly

With appropriate, time-limited access, we resolve the problem in your environment, document what we changed and why, and confirm it's working.

Best for: hosting, DNS, certificate and configuration problems.

Option two

We give you precise instructions

We diagnose the cause and write clear, step-by-step guidance for you, your developer or the AI tool you're working with — including how to verify the fix.

Best for: application code, teams who want to learn, or environments we shouldn't touch.

What to expect

Describe the problem. What's happening, since when, what's been tried, and how urgent it is. Please don't include passwords or keys.
Agree an initial investigation. We'll propose a time-boxed first step, the access needed, and the cost — before starting.
Diagnosis. We investigate and explain what we found in plain language, including anything else worth knowing.
Resolution. We fix it, or provide instructions — then confirm the result.
Prevention. Where useful, we suggest how to stop it happening again.

Questions about troubleshooting

Can you fix it, or just tell us what's wrong?

Either. We can work directly in your environment where that's appropriate, or provide precise written instructions for you, your developer, or the AI tool you're using. We'll recommend whichever is fastest and safest for your situation.

What access will you need?

Only what the problem requires — often read access to logs, hosting configuration and the relevant code. We'll explain what we need and why, use temporary access wherever possible, and ask you to remove it when we're done. Never send passwords or keys through the website form.

Is this an emergency service?

Not currently. We don't offer guaranteed emergency response times. Tell us how urgent the problem is when you get in touch, and we'll be honest about when we can help.

Application Troubleshooting & Technical Guidance