Software Rescue
We stop the bleeding: recover source code and access, fix the failures that wake you up at night, close the security holes, and put backups, monitoring and documentation in place. The system you have, made dependable.
[ Is this you? ]
A change that should take a day takes a month, if anyone will attempt it. The server has not been rebooted in years because nobody is sure it would come back. The business has quietly organised its week around not asking.
Fixed scope · 1–5 days · a report you own whether or not you continue.
When nobody will touch a system, it is because nobody knows what happens when they do. That is a backup and documentation problem before it is a code problem.
A one-day change takes a month because each step waits for the one person who might know, or for a window when failure would hurt least.
A system that has run a business for a decade has a working core. What it lacks is a way to change it safely: a copy to try things on, a test that says it still works, a runbook.
Nothing is rebooted and nothing is changed. The week ends with a copy that has been restored somewhere else and shown to work — the first time anyone has known it would.
We ask for the keys that let us look, not touch. Nothing on a running system is changed in week one.
Code, database, server configuration — a snapshot you keep, so there is always a way back.
Every server, repository, database and integration the business runs on, including the ones nobody remembered.
What is fragile, what is urgent and what it costs to fix, in plain language, with the evidence beside each finding.
The usual answer is rescue, not replacement: a restore that has been tested, a staging copy, the fragile pieces named and fixed in order. Once a change can be tried safely, the month becomes a day again.
We stop the bleeding: recover source code and access, fix the failures that wake you up at night, close the security holes, and put backups, monitoring and documentation in place. The system you have, made dependable.
The shipping platform had not one automated test; whether a change had broken something was found out by shipping. The plan gave it one-week sprints with something shown every week, and the old platform kept running throughout.
A shipping platform, rescued
Thirty minutes, no commitment. If a health check is the right next step, it is from $500, one to five days, and the report is yours whether or not you continue.