Engineering Management and Entropy

Throughout my time in Engineering management, I’ve had an evolution in thought about what a good manager does.

Initially when I became an Engineering Manager – I patterned a lot of my practices around the best that I had seen from better Engineering Managers that I had reported to in the past.

Fundamentally – the best practices I had seen have always related to both support and empowerment. I think that a good EM supports their direct reports by providing relevant context and removing details / abstractions that don’t matter.

Along with this, I think that a good Manager has empathy for what their reports do. They maintain this empathy by being able to do the same things that Engineers on their teams do – shipping and reviewing code, participating in discussions about design, and pushing their teams to be the best that they can – technically and product wise.

But eventually being a Servant Leader in this way (lead from the front) has diminishing returns. Why is that?

I think it’s because, fundamentally, after modeling behavior enough and showing your expectations the difficult thing about management is that you eventually work yourself out of some of your job.

At a certain point – your direct reports should know the technical surface area, understand implicitly what optimal behavior is on the team. They will rely on you more and more for higher leverage and more traditional EM activities, such as hiring, setting strategy and removing larger, more strategic roadblocks. They will be able to handle the rest of the tasks that they are best at – that they should be empowered to do because of your earlier, hard work. (and quite frankly, their hard work.)

Therefore, one of the tricky things about management is recognizing when it is time to rise to a higher mode of abstraction in running the team. It’s also the case that you might always be operating in a mixed mode of layers. Sometimes you will be spending more time with certain individuals on their career plans, but maybe less time on hiring. Then other times, you might have to dive back into a lower level and lead from the front when spinning up a new team. The art of management is recognizing the state of the system (where are things low entropy, where are things high entropy and highly tuned) and adjusting your style to adapt to the system and circumstances.


Comments

One response to “Engineering Management and Entropy”

  1. Nice

    Like

Leave a comment