Engineering Management, Fast and Slow
The two loops that drive the success of Engineering Managers
Last week, I got pulled into a huddle. Half my team was already there. This could be serious.
“Hey, we're looking at this issue. Service ABC is running into high disk usage.”
Just a typical day for an engineering manager.
In situations like this, you often face a dilemma. Do I jump in and help fix the issue, or step back and watch? One path gets quick results; the other builds long-term resilience. It’s fast vs slow. Simple, right?
Wrong!
In this post, I'll introduce the concept of Fast and Slow Management Loops — two interconnected approaches every manager should master. By the end, you’ll see why you need both to succeed.
We’ll think of a loop as a cycle of “situation → action → result.”
Fast Management Loop
The Fast Management Loop is all about quick actions that lead to quick results. You rely on instinct and experience to make decisions under pressure when waiting could cost you time, money, or trust.
Examples:
Crisis: A production outage or a critical bug.
Deadlines: Your team is struggling to meet an important milestone.
Blockers: A project is stuck due to external dependencies.
Slow Management Loop
The Slow Management Loop, on the other hand, is about making decisions that don’t show immediate results but build long-term stability and growth for your team and systems.
Examples:
Long-term planning: Crafting the team’s vision, roadmap, or culture.
Team development: Growing and rewarding your engineers.
Navigating complexity: Driving a complex or ambiguous project.
Only Relying on Fast Loops
Making decisions and solving problems under time pressure feels exciting, but if that’s your only approach, you may be:
Ignoring your systems: Quick fixes often lead to fragile systems.
Becoming a bottleneck: If you're always jumping in, your team will depend on you too much.
Hoarding knowledge: You’re solving problems yourself instead of helping the team to learn and grow.
Only Relying on Slow Loops
While the Slow Loop is deliberate and thoughtful, relying on it too much can also have drawbacks. You might be:
Lowering morale: Without quick wins, your team could feel stuck and unproductive.
Not addressing urgent problems: Critical issues could linger too long without immediate action.
Overthinking: Decisions may be delayed, and analysis paralysis can set in.
Fast AND Slow: A Dual Approach
Successful engineering management requires both Fast and Slow Loops working in tandem. You need to trigger a Fast Loop for immediate relief and a Slow Loop for sustainable growth.
Example: The Disk Usage Issue
I knew about the recent changes in this service and looks like something was missed. I needed to get my team out of this situation so they’re not demoralized as the situation gets worse. I also needed to make a note to dig deeper and prevent similar issues in the future.
Here’s how it played out:
Fast Loop:
I suggested cleaning up disk space and increasing the volume size to prevent immediate service disruption.
I paired two engineers to resolve it quickly.
Slow Loop:
I scheduled a root cause analysis meeting to figure out why the disk filled up so quickly, why the issue wasn’t caught earlier, and why there was no alerting.
I made a note for a 1:1 with the on-call engineer to discuss skill gaps and potential training since they weren’t able to solve on their own and needed to huddle.
Every fast loop you’re involved with is an opportunity to educate or delegate. Most of your value as an Engineering Manager, should come from the slow loops.
Exercise
I’ll leave an exercise here for you to apply the Fast and Slow loop model. Let me know your answers in the comments.
Scenario: Junior Developer Struggling
A junior developer on your team has been struggling to keep up with the workload and consistently submits buggy code. The rest of the team is frustrated because of this and often has to jump in to fix the issues.
Fast Loop:
Slow Loop:
More Resources
“Thinking, Fast and Slow” Book and Talk by Daniel Kahneman.
I based the concept of Fast and Slow on this book.
Firefighter vs Fire marshal article by
Loved the article Suresh.
Small question - in the long loop should an Engineering Manager also consider the employee turnover ?
You as an EM have invested substantially in building your team, what if this effort does not give you the result because they switch their jobs.
I am dont know if this should be considered at all by an EM.
i think the important that you figure out which one need to be solved first (fast loop). Both your given examples above I just faced 3 weeks ago, and I did exactly as you mentioned in this article.