Two and a half years at Swivt Technology taught me more than I had learned in three years of self-study. Most of it was not code. It was the messy, human, mostly-invisible parts of shipping work on deadlines for people who are paying for it.
Here are the notes I would give my 22-year-old self about how to ship under pressure without burning out — or being the person other developers dread inheriting projects from.
01Estimates are stories, not promises
When you say “this will take three days,” you are telling a story about how reality is likely to unfold. Reality has its own ideas. Bake in 1.5× to 2× of your gut estimate. The first time you do this it will feel dishonest. After 20 projects it will feel like wisdom.
02Scope is the lever, not effort
When something is slipping, the instinct is to work harder. Do not. The right move is almost always to renegotiate scope.
- “We can ship the editor on Tuesday, but the comments feature would have to slide to next sprint.”
- “We can include the third language at launch, or we can keep the launch date — pick one.”
This conversation feels uncomfortable at first. It is also the difference between a senior contractor and a junior one.
03Communicate in writing, even when you do not have to
Slack call this morning ended with “let us go with the second option.” Write it down. Drop it in the thread. CC the PM.
Three reasons: it forces you to articulate the decision precisely; it surfaces miscommunications before they turn into bugs; and six months from now when someone asks “why is it like this?” there is an answer.
04Saying no is a skill
“Can you also add a chat widget by Friday?” “Can you make the hero animate like this Awwwards site?” The reflex is yes. Yes is how you get respected and rebooked.
Until yes is also how you ship two weeks late and burn out.
Better: “I can do that, but it pushes the launch by N days. Want me to make that trade?” You are not saying no. You are saying here is the cost. Most stakeholders, given the cost, will choose differently.
05Rest is part of the job
When a deadline gets close I used to sleep less and code more. The bugs I produced at 2am cost more hours the next day than the sleep would have. Today I treat sleep as part of the project plan. The project does not fight me anymore.
None of this is unique to development. Every craft has the same arc — learn the tools, then learn the conditions under which the tools work. The second half is the one nobody teaches you.
If you are early in your career: be patient. The discipline takes a few years to build, and it is quietly more valuable than any framework you will learn.