If I was good at my job what would I be doing right now?
This is a question I ask myself when I’m in weird situations when I’m not sure what to do. Which is a thing that happens fairly often. The phrase gives me a way to recognize that I am actually not feeling great about myself, and gives me the space to try to do something different and better.
This is about how I try to work and is a reflection on the kind of software engineer I am, and will hopefully help you if you’re facing uncertainty as well.
(I'd be doing this talk.)
When I joined Datadog, my manager pointed me at the onboarding team and said, "go there and help, it's important."
The entirety of my onboarding experience with him: here’s a list of people, here’s a project that’s happening, go do that. We had worked together before so he did the responsible thing and threw me into the deep end. I was supposed to be able to figure it out, that's my job. I was new to the company, the team, the stack, the product, etc.
There's something forgiving about being new — you're given leeway to take your time. I got to sit in on meetings and not have to contribute much for a bit, I met a bunch of people, had a lot of conversations, learned things. I took a lot of notes.
The question was still there... "what do I do?", but first I needed to know: what was going on; and then how it was going on; and why it was going on.
I didn't want to get in the way of the work folks were already doing. And if I was going to help, I needed trust from the folks already doing the work, obviously that helps me trust what I'm doing since they're the ones who have been working on things, they're the ones who are the experts on the space.
I didn't have a lot of time, I knew that I had 3 months-- we were going to have another baby. So my goal was to make a tangible impact within that time period and if I didn't, I'd have done nothing for the first 6 months of my job and that would suck.
This is a harsh way to look at myself and feel but from the feedback cycle and evaluation and compensation and just feeling good about my time there, I would want to have had an impact and not come back and then start over entirely again.
When you start somewhere, it's the same, "is this person good?", "what are they good at?", "why are they here?", "what are they going to do?"
This is true for everyone new on a team, and new at a company and new on a project. When you start somewhere at a higher level, that expectation is higher, you have to have impact sooner.
"If I was good at my job what would I be doing right now?"
The first time I really articulated this thought to myself was in my last job, I was a new manager, managing a team that hadn't had consistent management. This was at CashApp, and it was the traffic team, just to be specific.
It was "well understood" they were really smart but the team was underperforming, but no one knew how they were underperforming, it was vibes, they were working on a lot of super important projects, and I was there to help them? I had no idea what I was doing. I didn't know how to manage a team into a better situation. I'd never helped in that way before.
I was stuck, I was full of dread and anxiety and I couldn't and shouldn't be doing any of the usual "engineer” things to help this situation, in some way, you have to help the team figure things out for themselves, but guide them there, wherever way that is so that they learn how to do that for themselves in the future — and that’s growth. But you have to do that understanding the general power dynamic of being their “boss.”
"If I was good at my job what would I be doing right now?"
After some time fighting myself, I got here. I knew that there were people out there that were good at this kind of thing, in fact, I’d been on teams that were in similar situations, and needed help, and I might have even helped, but this was different.
What would someone who knew how to solve this problem do?
It gave me permission to admit to myself that I wasn't doing a great job. Even if it was OK from the outside because I was a new manager, I didn't feel like I was doing a good job, and that could kind of get me to forgive myself, and give me an outlet for that experience.
What does the team need? What do the stakeholders— what do the folks who think the team is not doing well need? How would someone else solve this problem? Can I do a little bit of that? Then can I do more of it?
Of course, the answers eventually were something like: more communication, stop working on things you don't need to work on, get the team to stop doing things that were contributing to this rut they were in, in a way that helped them learn they could do this, as well, which is also what I was trying to do for myself at the same time.
I realized that even before the CashApp stuff, and before I had this question to help myself, I was kind of already doing it. It just wasn't intentional or conscious.
I used to work at Etsy, I was there in total for almost 7 years. 5 years in, The CEO was fired and there were a few rounds of layoffs. And when you have layoffs, other people leave too, so obviously there were fewer people, and more importantly, there were fewer people that did specific kinds of work. I did not leave.
Later that year we had a team of product engineers working on a feature that hadn't worked on anything like that before— they were transferred from one department to work on the user facing product. I was asked to help them— there were a few weeks before the deadline to ship— it was a holiday promotion, so like a real deadline not an end of quarter announcement that can say “coming soon”.
I remember thinking that there were two bad outcomes:
I had context on how to build products at Etsy but not how this team worked, what they were building, and what they needed help with.
I also didn't have their trust. I was someone forced onto their team — a team that had been working together without me just fine.
The one thing that immediately gets dropped from scope when you're under a tight deadline is tests.
I wrote a ton of tests, and that ended up saving a lot of time because the team would have to manually validate their changes every time they made them.
Tests give an engineering team confidence to make changes and move faster. It also helped me understand what they were doing, and influence their behavior with a different way than just saying, "I think we should do it like this because I’m senior."
In a way I did their “unfun maintenance” and used that to make their work “more testable” and then more robust, and continue to slowly introduce patterns/changes. So they could both work faster since they had to do manual testing less often, and the quality of their work went up since the code was now safer to change.
They built the features. They shipped the product. They succeeded as a team. They won. They got recognition. I just helped.
Here we are, back in the future, at Datadog, APM Onboarding and Single Step Instrumentation.
Fast forward six months into Datadog and there’s a new team in place! Injection Platform, with only one person remaining from the squad — with everyone moving on to other projects at the company.
I thought that when I went on leave — I left all of these ideas that would hopefully be worked on, and I would move on to something new — but Kubernetes didn't move. Linux is about to GA, Kubernetes is just hanging around. It has to GA as soon as possible. There is no plan.
Even though I thought my job was originally to do some productive things and then move on to another part of the org to kind of onboard and learn more, the thing that needed to happen here is a plan of execution. It was pretty obvious, I had the most context, it was a nice and natural fit, I didn't fight the inertia.
The big problem was there was no one on the team working on it, they hadn't hired someone for that yet, and in order for this to work we needed a plan and buy in —
The team was going to own this long term, not me. No individual really owns a piece of software at a company.
I knew that working on this and shipping to GA was going to be a thing I did right now not necessarily the space that I was going to occupy forever, I shouldn't be the kubernetes-single-step guy — we could hopefully get another one.
We hired someone, in NY, my manager told me he knew k8s, and I kind of stole him to shadow and work with me on the features— we needed someone from the team to own the feature, and be setup to own the general space longer term.
Luckily, he was great and he sponged everything up as quickly as it was available, was able to deliver the work and now is the person on the team who owns all of this.
I stopped writing code on the project except for small things — he shipped the big features, he worked across team with the container platform, he got it done. Him and the team succeeded. For me, a sign that I'm doing my job well, is that other folks succeed and I don't need to be there anymore.
My role shifted from “earn trust with a new team" to “lead new team member to own the space” — from implementation to coaching and review and design to trying to step back and make room for somebody else.
And then I was in limbo.
A pattern I've noticed is that this is a mildly compulsive and anxious behavior for me, I've been doing this for years and can finally describe that I go into these cycles of certainty and uncertainty as a function of being me.
I’m continuing to put myself in a position as an impostor. I'm doing this to myself. Parts of this are very much not fun. I give up immediate expertise, I give up immediate productivity, I give up the gratification of being able to ship features and gain and keep momentum, it just sometimes kind of stops.
No flow state.
The positives are I work with a lot of people and unlike being a contractor, I get to keep these relationships, I stay at the same company. It's a lot of discomfort traded for gaining context and expertise at more of Datadog-- which compounds over time.
But yeah, I was still in limbo. My manager asked me to check the service remapping project— super ambitious, super important, huge in scope. I showed up a stranger again. With a bunch of people who already had working relationships and trust. I needed to figure out how to fit in.
There were great people who knew the domain deeply, and there was already senior engineering leadership from a prolific product engineer who has been at Datadog for a long time — who knows a ton about literally everything. Again, there were vibes. There was uncertainty coming from how folks outside of the squad were looking at the effort, and the team needed full support from engineering and product leadership to move forward with the plan because of how massive an undertaking it was.
The gap was clarity between the team and leadership — everything else was there as well as it could be. I put together a document with the team that coalesced the risks and plan that was based on the work they were already doing. It was a nice doc. That was it. They didn't actually need me. Anything I would add might actually be a detriment to the success of the project. The most helpful thing I could do was stop helping. So I stopped.
This was hard because, again, I didn't have a “thing to do.” My boss had just asked me to work on maybe the most important thing at the time, and I didn't.
When I talk about these ups and downs, the downs can still feel like despair. I've been through these cycles a lot so it just gets a little easier. It was my wife who once actually looked at me and said:
"So you’re having a hard time right now? Okay, let’s check on how you’re doing in a few weeks, pretty sure this happens a few times a year. This is normal for you-- when are you going to figure this out?"
There are a few things out of these examples (and there are more) that I’d want for folks to take away.
If things aren't going well right now, that’s just a thing that’s happening right now. We do this in post-mortems, retros, etc, but this is true for introspective things as well.
Thinking about what other people are really good at is a fun way to think about what you’re good at, too.
If you’re trying to figure out what to do or how to grow, sometimes it’s by looking around you and seeing what the people around you are really good at and trying to copy that.
You can also see if there’s something that people around you aren't doing that feels like a gap and step into that. You don't have to be perfect at it. No one is doing it!
You can help support your teammates in many ways. With docs, or tests, or just having a conversation, or setting up that meeting to discuss the thing everyone has been avoiding.
Doing nothing is a choice and even though it feels like a default. When you realize that you can choose to wait until you have a better option than what you have now, it’ll give you the space and time to see it. First, do no harm.
Seeing what someone is not doing is hard.
How is equally if not more important than what. It’s not just what, it’s how. And how you approach work models how other people approach their work. Curiosity, empathy, trust — these are things you show people through the work you do and hopefully encourage the same kind of behavior in the people around you. How is what you show when you show up.
If I was good at my job how would I be doing it?