Forward deployed and solutions engineering

Stage 2 · Lesson 6 of 6

Debugging systems you did not build

Core engineering you still need

At a customer you rarely know the system, and people are often watching. A calm method looks competent and works: reproduce the problem with the smallest case you can, narrow it down by halving where it could be, form one hypothesis, test it, and repeat.

Read and traces for the request ID and the time of failure, since logs are usually the only witness to something that happened overnight. Write down what you tried as you go, which also becomes the incident notes. Say out loud what you are checking, so the customer sees progress.

Common trapchanging several things at once under pressure, so that even when the problem disappears you cannot say why, and it comes back next week.

A question to check yourself, an AI practice session and 2 cheat sheet entries come with the full roadmap. Unlock

Resources for this stage

Start with· follow one of these

Practise

  • Fifty graded SQL problems to build speed with joins, aggregates and window functions

Keep handy

  • The clearest reference for HTTP methods, headers and status codes

  • How to send and inspect any HTTP request from the terminal, the first tool in integration debugging

Also worth your time

Mark this stage as not done?