Being an Engineering Manager at Meta
3 lessons from 3.5 years at Meta
4 months ago, Gilad Naor shared his lessons from his time as an Engineering Manager at Amazon.
I really enjoyed collaborating and learning from that article, so I invited Gilad for a 2nd round - sharing his lessons from 3.5 years as an Engineering Manager at Meta!
Gilad recently launched a free(!) course, titled “The Essence of Management”. If you want to level up your management career, and learn from someone who experienced it all (startups+big tech), check it out:
Not sponsored!
Today Gilad is going to share:
3 lessons he learned from Meta
What he doesn’t miss about working at Meta
What he misses the most
Passing the mic to Gilad!🎤
#1 Engineers First
“I don’t care about managers, I only care about individual contributors.”
That’s what my Senior Director told us.
At Meta, managers do not manage a team. At Meta, managers support a team.
This is not just how people talk. This reflects the actual ingrained management culture. Meta has a strong bottoms-up culture. Software engineers are expected to choose what to work on. More senior engineers are expected to set the direction for their team and organization. This expectation is literally one of the four performance evaluation dimensions: “Direction,” along with Impact, Engineering Excellence, and People.
As a manager, this was a mindset shift that took some getting used to. The bottoms-up culture happens in parallel, not in place of, the top-down direction. Part of the challenge of being a manager at Meta is connecting these two approaches. For example, an engineer may want to build a new feature that will increase operational costs while the top-down goal may be to increase efficiency.
An interesting aspect of this is that engineers are free to pursue challenging projects and set their own goals. As a manager, I had to clarify what the company expectations were and work together with them (and sometimes data-science) to align the goals with the expectations. But at the end of the day, if the engineer felt strongly enough about something, they had the freedom to pursue it.
The Tip
Your engineers are capable of more than you think. Trust them. Choose a small business outcome and ask them to achieve it. Give them ownership and autonomy to figure out their own path. Hold them accountable and support them. You will be surprised.
#2 It’s All About the Deltas
Managers at Meta are evaluated based on four main dimensions. The first two are the most important: how much Impact their team landed and how much they grew their people.
There are a lot of misconceptions on how this is evaluated both internally and on Blind. People believe that managers get credit for what the individual contributors on their team did.
Here is how people imagine it plays out.
A strong engineer comes up with a new project. They build it, launch it, and much Impact follows. The engineer did such a great job that they also got a promotion out of it. Now people imagine that the team impact and people growth dimensions of their manager will look great and the manager will get a great rating.
It doesn’t work like that.
The most common question during performance calibrations for managers is: what did the manager contribute? Or, in other words, what would have been different if the manager wasn’t there?
Now it becomes a question of coaching and support. What mentoring did the manager provide to the strong engineer that helped them succeed? Who did they connect them with that helped them identify the project? How did the manager challenge and stretch the individual? Did they help them hone in on their strengths and leverage them further?
After years of management, this was an eye-opening experience for me.
I actually believe that this is the most important mindset shift that a manager has to take. This is the main reason why I invested so much in creating a free course so that more people can internalize this shift.
The Tip
At the end of each week, look back at what you accomplished. Think about all your small contributions. Hallway conversations that led to you saying: “You should really talk to Jenna about this.” Fast in-the-moment feedback that you provided. The act of documenting these contributions will make you better at noticing more opportunities to help.
#3 Move Fast. REALLY Fast.
It took me less than a minute to get a complete development environment ready.
I was in Meta’s Bootcamp program - the company-wide multi-week onboarding program. My job was to make a change to the code base that was publicly visible. It was in a language that I never worked with before called Hack. There is a good chance that you have never even heard of it before. I sure didn’t. And yet, here I was, landing a code change in production in my first week at the company.
In contrast, getting a development environment at Amazon took me nearly an entire day. I had to follow a detailed wiki page detailing the various manual steps that I had to complete to get my environment set up. The wiki page included lots of branching “goto” statements. If you are in this org, skip to this section. If you are in that org, skip to that section.
One of Meta’s long-standing values is Move Fast.
It is also the most misunderstood value.
People think that it is about breaking things. Or that it is about cutting corners.
Meta’s amazing developer experience didn’t happen overnight, and it is not cheap. Most of the tools are developed in-house, including the most common programming language/compiler/runtime (see Hack) and source control system (see Sapling). This is a long term investment that is made so that the company as a whole can Move Fast.
The Tip
Look for opportunities to move faster. There is a good chance that you are thinking about reducing Tech Debt. But will reducing the technical debt really help? You need to have a clear narrative of why making an investment will actually enable you to move faster. Can you explain why in 2 minutes?
What I don’t miss about working at Meta
Meta pays a lot. What many people don’t know is how much Meta pays top performers. Even without promotions, strong performers can double and triple their overall income at Meta.
In parallel, Meta invests a lot in identifying who the strong performers are. As a manager, this Calibration process requires a lot. It requires a lot of time. It requires a lot of emotional ups and downs. The process is not perfect, but it does try to maximize fairness in ratings.
I actually miss Calibrations. It was weeks of working late and making compromises in my personal life. It was emotionally draining to have the people on my team challenged by managers and Staff+ engineers from other teams. But I also learned a lot and the overall goal was noble.
So what do I not miss about the Performance Summary Cycle (PSC)?
I do not miss how much it bled into the culture.
It often felt that people optimized work for their PSC rating. If some work wouldn’t count, they just wouldn’t do it.
What I miss the most
Meta is a social company at heart. The products that they build are all about human connections. The company is all about human connection.
I miss how open and freely people could connect within Meta.
That Calibration process that I mentioned?
You literally get performance credit for building connections within the company. If you are passionate about Vipassana Meditation and start an internal club that takes off, you will get credit for it.
This contributed to the most open and welcoming culture that I ever experienced. I could talk to anyone in the company. Just throw a meeting on their calendar, and they will show up. Or, if I wanted to learn how to make coffee, I could join the local club and meet people from all over the company.
As humans, we are social creatures. We evolved to work in groups. I honestly believe that companies that try to be a “family” are aiming in the right direction but are still wrong. A company is not a family. But neither is it a bunch of disconnected humans. Meta gets this right: a company is a group of humans working together, forming real connections with each other toward a common goal.
Final words
Anton here - thanks Gilad for another great guest appearance!
This is the third article covering Engineering Management in different companies, after Amazon and Roots. In 2025 I’m planning to have ~10 more of those, currently have one in progress with an EM from Miro!
If you work at an interesting company, and want to share your lessons from it, ping me here or on LinkedIn :)
What I enjoyed reading this week
 | A guest post by
|
Great article, guys! I’m curious about the moving fast philosophy. While I was working on my startup, where we created custom software for clients, we always took the time to write clear onboarding documents and set up steps for the team working on the project later. But these were smaller projects, so I thought that’s why anyone could get it running in hours.
However, working now in an enterprise with a more complex project, I assumed that it was normal for me to need a day for setup.
But now reading your experience at Meta I began to think, maybe it depends on how much effort you put into developer experience? I’m sure whatever you set up at Meta was more complex than the current enterprise project I work on.
I just had a question, the goal of a manager (eventually) is that you grow your people enough that they can autonomously go from 0 to a 100 in a project with little to no oversight. Your contribution as a manager will be smaller and smaller but that, to me, means you did a great job. When a great IC performs very well (like in your example), is that accounted for? You might actually have little to no contribution for that project, but that doesn't mean you didn't support and help then get to that point.