tushar

The Unsexy Habits That Make You a Great Engineer

Nobody writes blog posts about showing up on time, or reading error messages, or actually doing what you said you'd do. But these habits more than any framework or language are what separate engineers people love working with from those they don't.

Great engineering is largely invisible. It's not the 2 AM heroics that save a production outage (though those matter). It's the hundred small disciplines practiced every single day, the ones that compound quietly into a reputation for reliability and craft. Here's a collection of those habits, assembled without apology for how obvious they might seem.


Testing & Quality

Test it manually at least once, before review.
Even if you wrote thorough automated tests, sit down and use the thing yourself before asking a colleague to review it. You will be surprised what you find.

Actually test that your test would fail.
Every time you add a test, comment out the code you think makes it pass. Run the test. Watch it fail. Comment the code back in. Run it again. See it succeed. Only then have you written a test not just a permanently green fixture.

Think through the edge cases.
You don't have to handle all of them immediately. But you should have thought about them and have a defensible answer ready when someone asks. Blind spots in edge case thinking are where bugs live.

Know why your fix is a fix.
Not just that it works but why. If you can't explain the causal chain between your change and the resolution, you may have masked a symptom. Dig until the explanation satisfies you.

Read the error messages.
The computer is usually telling you exactly what went wrong. The number of hours lost because someone didn't read the message in front of them is a tragedy of the profession. Read it. Read it again.


Code Reviews & Collaboration

Keep your change free of unrelated changes.
Every unrelated change in a PR is a distraction for you and for the reviewer. It conflates intent, muddies the history, and makes rollbacks harder. Scope your work tightly.

Keep your PR up to date against main.
Don't make reviewers mentally calculate drift against a stale branch. Stay current. It signals respect for their time, and it forces you to deal with conflicts before they become crises.

Read the whole thing before you comment.
This applies to code reviews, documents, tickets, and threads. Comments that would have been answered two paragraphs later waste everyone's time. Before you comment, finish reading.

Don't ask for "quick" reviews if you never review others' code.
The social contract of code review runs in both directions. If you want fast, thoughtful feedback, you have to give it too. Reciprocity is not optional.


Professionalism & Mindset

Do the homework before the meeting.
Know the agenda. Know the context. Read the document that was shared. The person who arrives prepared stands out immediately because it's rarer than it should be.

Read the manual, the docs, the instructions, the ticket.
Before asking, check if the answer is already written down. Often it is. The discipline of reading first respects the time of everyone around you, and almost always makes you faster.

Be on time.
For meetings, for deadlines, for standups. Lateness communicates that other people's time is worth less than yours. It's a small thing with outsized effects on trust.

Do what you said you'll do or communicate if you can't.
Reliability is one of the rarest and most valued traits on a team. If something comes up and you can't deliver, tell the person who is counting on you, early. Don't let them discover it themselves.

Accept that some of the work is just work.
Not every task will be intellectually thrilling. Some tickets are tedious. Some refactors are unglamorous. Accepting this without resentment is a mark of maturity. The unsexy work done reliably is what earns you the interesting work later.

Write clear, useful bug reports.
"It doesn't work for me" is not a bug report. A good one tells the reader what you did, what you expected to happen, and what actually happened. Bonus points: find the minimal set of steps to reproduce before opening the ticket.

Always keep an eye out for bugs.
Your job doesn't end at the boundary of your feature. If you notice something wrong while working on something else, surface it. A culture where everyone is watching makes the whole system more reliable.

Make it a goal that people want to work with you.
Technical skill is a foundation. What multiplies it is how you show up for others whether you're reliable, whether you communicate, whether you give as much as you take. That's the real career.


None of these habits require genius. They require attention, consistency, and a genuine respect for the people you work alongside. The engineers who master them aren't just productive they're the ones everyone wants on their team, trusted with hard problems, and who quietly shape how a team's culture feels.

Do the small things right, every time. The rest follows.