Short answer
The first year of engineering management is hard because the feedback loop disappears. You stop shipping code and start being measured on other people's output, which arrives weeks later and noisily. Most of the job is a small number of conversations done well - and those can be practised.
You were promoted because you were a strong engineer. Nothing about being a strong engineer prepares you for telling someone their work is not good enough, or managing the person who applied for your job, or ending a week having personally produced nothing.
The Missing Feedback Loop
Engineering gives constant signal: tests pass, the build goes green, the feature ships. Management gives almost none. A good hire pays off in six months, good coaching in three, and a bad decision might not surface for a year. New managers routinely interpret this absence of signal as evidence they are failing, when it is simply what the job feels like.
The conversations that make up the job
- The 1:1 - the core unit of management, and the one most commonly wasted on status updates that standup already covered.
- The performance conversation - the one new managers most reliably do badly, almost always by being too soft to be understood.
- The former-peer conversation - awkward, unavoidable, and much worse if you postpone it.
- Saying no upward - protecting your team's scope from commitments made without them.
Start here
Your first 1:1 has the agenda, the three questions that get real answers, and the script for the report who wanted your job. Run it in your first week.
The performance conversation is the one worth reading before you need it. The most common failure is not harshness - it is being so gentle that the engineer leaves believing everything is fine, and then being genuinely blindsided months later.
On imposter syndrome
Almost every new engineering manager feels like a fraud in the first six months, and the reason is structural rather than personal. You have left a role where you had daily proof of competence for one where proof arrives quarterly. Imposter syndrome in a new manager role is about changing what you count - obstacles cleared, decisions unblocked, people who grew - rather than trying to feel more confident.
Should you have taken the job?
It is a fair question to still be asking in month four. The engineer-to-manager transition lays out honestly what changes and what you give up, including the part most advice omits: going back to individual contribution is a legitimate, common and often career-enhancing decision rather than a failure.
Frequently asked questions
What should a new engineering manager do first?
Run 1:1s with everyone in the first week and mostly listen. You have positional authority and very little information, and the licence to ask naive questions expires quickly - spend it before it does.
How do I manage engineers who used to be my peers?
Name the change directly and early rather than pretending nothing happened. One uncomfortable conversation acknowledging the shift prevents months of ambiguity, and it is especially important with anyone who applied for the role you got.
Why do I feel like a fraud as a new manager?
Because the feedback loop that made you feel competent as an engineer has gone. Management signal arrives in months rather than minutes, so the absence of daily evidence gets misread as evidence of failure. It typically takes about six months to recalibrate.
How much should an engineering manager code?
Enough to stay technically fluent, but never on the critical path. Design reviews, small non-blocking fixes and reading code work well; owning deadline-bound features does not, because management work will always preempt it.