At our recent online webinar, Career Moves for Senior Developers, we interviewed both Jeremy Edberg (Founder of MinOps and former Architect at Netflix and Reddit) and Ray McDonnell (Enterprise Account Director at LinkedIn).
Read the full interview below, or watch it from 42:35:
Interview with Jeremy Edberg
Stephen: [starts at 42:35]
First of all, like I said, thanks so much for continuously giving back with OfferZen. You’ve done a lot of stuff with us and spoken at a few of our events. Thank you.
I think the first question for me is to set the stage and ask you to tell us a bit about how you jumped into your career as a software developer: the very high-level route you’ve taken and how you thought about it?
At no point in my career could I have told you what I was doing in five years. I kind of just go with the wind and find opportunities. My first opportunity was working for the university where I was going to school. They had a job application at the computer centre and I ended up working for them my entire college career. A year after that, my friend who worked at a startup said “Hey, this place is really cool. You should come here”. And I did.
For Reddit, I went to a startup school. I met the founders, they were cool people, and I just said to them, “Hey, you guys are cool, and I like your product. I want to work for you”. They were like, “Well, as soon as we have some budget to hire, we’ll let you know”, and six months later they messaged me and said, “We’ve got a budget to hire now, so come on by”.
The same for Netflix. I was messaged out of the blue from the person who ended up becoming my manager because he saw my blog post on Reddit about how I was leaving Reddit. So I thought, “Well, if I got multiple angles coming in from Netflix, I might as well give it a shot”. It hasn’t always been rosy: It happened at Google, I didn’t get accepted there. Same thing happened at Facebook.
How do we distil that knowledge? Is it basically just you know people, and do things?
I think the main thing there is
- Build your network: Most of my best opportunities came from my network, telling me about them.
- Seek out opportunities: Reddit was not in my network. I specifically went to startup school with the intention of meeting those founders, talking to them, getting to know them, and trying to get a job from them. It wasn’t really planned, but it was intentional when the opportunity arose.
I think that’s an interesting point for senior software developers. There’s a lot of choices, and everyone is getting inbound messages from various platforms. Do you have a way of thinking about these things? What are the things that you’re prioritising, and how do you navigate them?
For me, I’m prioritising mostly on project opportunity. Will I be able to work on something big, cool, interesting, and new to me? I’m always looking for something that expands my skill set, work on something new, something different - things like that.
Yeah, I think that sits nicely with the data that we just talked about that senior devs are really looking for challenging projects. Tell me about Reddit. You started super early there, right?
Yeah, I was the first or second employee, depending on who you ask and how you calculate it but very early on.
That’s massive scope, like it’s just an open field.
Well, not then. It definitely wasn’t massive in scope. We were pretty small, but it was a great growth opportunity. It was a chance to really lead the tech from the beginning. I was basically the only one in charge of how we did any sort of DevOps. We didn’t call it that at the time, but that’s how we operated. All of our deployments, everything like that, I focused on the System Administration. I did some coding there as well, but mostly I focused on coding for System Administration.
Right. I’m trying to figure out a little bit about your mind, you start as a junior, then intermediate, then senior in your career. Do you have points in your memory where you were like, “Okay, I think I’m an intermediate now… I think I’m senior now”. Were there points?
It’s interesting. When I got to eBay, I was listed as just the engineer, not senior principal - just straight engineer. I did that for a few years, and I kept getting better and better, and then the promotion actually came literally the same day that I turned in my resignation.
I handed over my resignation, my two weeks’ notice, and my boss handed me a paper giving me a promotion to senior engineer, so that was where it stepped up. Then I went to Reddit and we were all principal engineers because we were all in charge of huge chunks of the application. I would say Reddit was where I skyrocketed from senior to principal, because when I left Reddit and went to Netflix, I was effectively operating at a principal level at that point.
Okay, amazing. You’ve also picked different companies - you did startup school and Reddit at that point was super start-uppy. When you got to Netflix they were not start-uppy. It was in scale mode, right?
Netflix was really interesting. They had been around already for 10 years, but at that point, they were just getting into streaming. Streaming was only maybe a year old at that point, and they were just really ramping it up. It was sort of a start-uppy in that sense, but also had tons of resources because they were a big public company, had been around for a while and had a big chunk of revenue from the DVD business. There were maybe 1000 people when I joined Netflix.
When you joined, were you in charge of a team that you worked with?
So that’s funny, when I joined there was supposed to be a manager who joined on the same day to be my manager, but that person ended up backing out at the last minute to go back to their old job, so I ended up being the Team Lead without intending for that to happen.
I ended up being the Team Lead for the reliability engineers. We eventually hired another manager, they came in, they left, another manager came in and left, but the whole time I was basically the Team Lead for that. So yeah, I worked on hiring and management the whole time, but not in an official capacity.
How did you experience that? Because that was your first shift from an individual contributor to a lead management type role?
It wasn’t because, at Reddit, I was basically the only engineer who had been there for more than a couple of months since everyone else had left. I was the default engineering manager at that point as well. I ended up hiring a team of five people right before I left, and I had done some management work. Actually, even way back at university, I had been doing management work for about a year. I was managing a team of five. I had five direct reports as well as 25 indirect reports, so I’ve had management experience throughout my career.
You’ve definitely lived it in different shapes and forms. One of the interesting things, when I was looking at the data, was that people move into a team lead and start thinking about their careers a bit differently. With career progression, once you are a senior (and you bring in the concept of principal, engineer, or lead), many people think that the end of a career path for a senior software developer is management. What is your take on the path of a Software Developer? How does it fork, or where does it end?
I agree. Management is the end of the path. I mean, it doesn’t have to be. I have friends who are managers who still write code occasionally, but the higher up you get, obviously, the less you can be doing, and you really shouldn’t be in the main line of fire . You kind of have to go there, and there’s this whole myth about, we have a tech track, that’s the same as a management track.
But that’s kind of what I was asking - is it a myth?
That’s turned out to be a lie. Yes, it’s true. It exists that there are some people at some software companies that get VP level pay as individual contributors, but you know, it is those people because they are world-famous. If you want to be at VP level pay, you need to be the best in the world at what you do. If you want to be a VP level manager, you need to be a really great manager.
Okay, and in that transition, it sounds like you’ve gone backwards and forwards through managing and leading. What do you think your biggest takeaways from that were, and what is the lesson that you will come into if someone is about to go into Jeremy’s career path? What would you tell your younger self in that space, having to navigate all of that?
I have enjoyed my career path. It’s been great. I’ve ping-ponged between a bigger company and a startup. It’s a different type of experience. I really enjoy leading, managing and mentoring people. That’s my favourite part, the mentoring and teaching, which is why I love to do stuff like this. I’ve definitely leaned more towards the management side and less towards the actual coding side as I’ve progressed in my career, because that’s what you have to do if you want to progress.
I have a question from someone, Alex Steinberg, who sent in the question beforehand, which I think is probably relevant for now. He asks, “When do you specialise and go deeper and when do you generalise and learn other parts of the stack?”. Alex, says that he generally follows the intersection of his interest and the needs of the company, but he’s wondering if you’ve got some wisdom for us.
I would say I always try to learn the stack when I get to a new company just because I want to know how my piece fits into the stack. You know, when it came to Reddit, I needed to know the stack because I was working on the entire thing.
When it came to eBay, it was interesting to learn about these things. I would seek out opportunities and projects, which allowed me to learn other parts of the business because I just wanted to learn other parts and how I fit into it.
With Netflix, my job required me to understand how all the pieces fit together. So that was sort of the job, but I like to learn all those pieces of the stack because I like to learn how they fit together. Oftentimes, when I’m working on something, I can find a way to make it better by knowing “Oh, hey, this other piece does this”.
So, you’re saying you can do better in the piece of your stack by knowing the rest of the stack better?
I would say so. Yeah. But you can’t spend all your time learning the rest of the stack if you still have to do your job or your part. It’s a balance.
I’m going to say no real wisdom other than balance it.
I think the real wisdom there is to seek out a project that requires you to learn other parts of the stack. Seek out a project that incorporates knowing other pieces.
I think that’s part of seniority and increased responsibilities. Understanding that we’re being responsible for the integration points in the stack is, by definition, increasing responsibility.
That’s a bonus. You’ll build your network because you’re gonna have to meet new people and talk to new people about other stuff.
Right, so increased horizontal knowledge and increased network, even within a company, depending on the size. One of the things, when we were chatting ahead of this, you pointed me towards a video of you talking about Python in the cloud back in 2010, and 11 years later, we see that Python is the most sought-after language. My question is, do you consider yourself a visionary?
I consider myself lucky.
The serious part of the question is, as a bunch of senior developers on the call, we need to think about where we are headed. Where do you think we are headed and where should people start thinking?
The reason why Python is growing again is because of AI machine learning and data science, so Python has amazing data science libraries. These things tie into machine learning and AI. Everything is starting to get some bit of machine learning. Then machine learning is just becoming a general term for using statistics and math to make determinations.
I think that’s kind of where we’re going. This has been the case forever: Automation. You can’t get to where you want to be without automating most of what you’re doing. When I was doing reliability at Netflix, we always talked about how the only way to get to five-nines of uptime is if everything is automated.
Remediation needs to be automated, and the way you do that is through data analysis and coming up with automated ways to do remediation. I think that’s kind of where we’re going is automating everything by using statistics, data science, machine learning, actual AI things.
One of the other questions, which is the top question on the actual report, if you go over to the Netherlands report - was, what do devs think will be the most interesting or promising industry in the next five years. And AI is at the top. It’s kind of surprising, and I guess your Python meets data religion thing is true.
Interview with Jeremy Edberg and Ray McDonnell on Career Moves for Senior Developers: Data Deep-Dives
We’ve got a bunch of questions coming in from the audience, so I’m going to invite Ray back so we can chat with him as well. While he joins, the first question is for you Jeremy, so I’ll ask it in the meantime.
Richard wants to know, question for Jeremy,** what are your suggestions for juggling side projects while working full time?**
When I didn’t have kids, it was a lot easier because I had plenty of time in the evenings to do that kind of thing. Now that I have children, they’re basically my side project, but I still have some side projects. I tend to do them late at night and on weekends. It’s not easy. You have to sacrifice something else - going out, watching a movie or whatever. Hopefully, your side project is so interesting that you don’t consider it a sacrifice. You’re excited about working on your projects instead of watching, you know, Godzilla versus King Kong.
Tell me about your Jeopardy Bell side project really quick. I think that’s a really cool story.
That wasn’t much of a side project. It was literally just programming my echo devices to make an announcement throughout my entire house that Jeopardy starts in three minutes, which is a way of balancing my work life because where I live in San Francisco, Jeopardy starts at 7p.m. The alarm goes off through the whole house and if I don’t come out within one minute, my kids are coming in here to remind me that it’s time for dinner and Jeopardy. It’s a way to force me to stop working at 7p.m. every day, no matter what’s going on. I have to pretty much drop at seven o'clock because they expect me to be out there at seven o'clock.
Self-enforcing your work-life balance.
Yeah, yeah, exactly. We’ve used the echo for a bunch of stuff, like to get my daughter to school on time. She’s doing remote school this year, and we don’t have to walk her or drive her anywhere. So we just set up a bunch of echo alerts that tell her you should be brushing your teeth, you should be getting dressed now, you should be sitting at your computer now. We get to sleep in because Alexa is the one who’s telling everybody what they need to be doing.
Okay, we have a question for Ray. The question is, what are the most sought-after soft skills, and how do you think employers are evaluating these soft skills?
Yeah, great question. I’ll start with the evaluation piece. So how employers evaluate soft skills. It’s generally true, it’s a competency-based interview, so “Can you tell me a time when you collaborated”, or “Can you tell me your time when you had to persuade somebody, from one opinion to another opinion”. Oor they would always say a situational type question like, “What would you do if?” What we’ve seen over the last 12 months is that creativity, persuasion, collaboration, emotional intelligence, and adaptability, have been the skills that have risen up to the fore on the soft side. Particularly over the last 12 months, given what we’re all going through.
Okay, I’ve always wondered about the merit of these two questions. The one is, “Tell me about a time you did x”, and the other one is, “How would you do x”? Any preference between those if one had to choose?
Well, “tell me about a time when” you’re telling somebody your experience of when you’ve done something, and my little hack around that is that I personally use the star method when I’m answering those questions. It’s situation, task, action, and result. That’s what I go through when I’m answering those questions, and sometimes, I even shorten it, and I just go, here’s the situation, here’s the result because you have the tendency to waffle or elaborate too much during your interview.
We have another question from the audience for Jeremy. How does work history compare to having a demonstrable GitHub portfolio in terms of value and looking for new opportunities as the employer is going to take a good look at these companies versus juggling all these things?
I will give bonus points for having an active GitHub, but I will not penalise you for not having one. Not everybody has a GitHub. Not everybody has time for public side projects. At times, for some, their work has to be private, so I’m not going to penalise someone for not having one. It definitely helps if you do have one and if I can go there, and I can see that you have a project that’s active, or you’ve made some good contributions to another project, or you have something with a lot of stars, that’s definitely going to be bonus points.
I personally don’t have much of a GitHub. I usually will share little projects that I’ve done or useful tools or utilities on my GitHub, but that’s about it because most of my coding has all been either for stuff at home that isn’t really shareable or for work, where I can’t share it. I mean, Reddit code is open source, and you can see my contributions there.
But not everyone has commits on Reddit that they can post about.
Honestly, I barely have any posts there as the stuff I did was all in the private repo. It was all anti-fraud, anti-cheating, and redeployment stuff.
Okay. Another question for you, Jeremy. Obviously, you’ve worked at some big companies and had some serious amounts of responsibility on your shoulders. How do you deal with imposter syndrome?
You kind of just get over it, I guess. I’ve always felt like, “how am I responsible for this?”, but once you just start getting into it, I guess it just kind of goes away. I never actively tried to get over it. It just sorts of went away after a while.
After all, I do deserve to be here, and people seem to respect what I’m doing, and it seems that I’m doing the work well. I would say that went away when I got to Reddit because there wasn’t anyone to worry that I was posturing. It was just a few of us there. It either worked or didn’t, right. If I were an imposter, it wouldn’t work.
Yes. Results and a high trust environment really helps.
Yeah. Yeah, that’s a good way to put it.
Cool. Another question for Ray, and I’m really excited for the answer. Does LinkedIn data show any insights on work from home adoption since lockdown started?
Yeah. It’s a great question and super topical. Microsoft has thought about this a lot, and LinkedIn has talked a lot about this a lot. Even before the call I shared our rescue week with Stephen: where LinkedIn is giving everybody a week off work.
One of the factors behind that is that people are exhausted. People are working much longer hours, they’re always on, your home life and your work life lines have never been more blurred. I’m lucky enough to work for a company where they really put a huge focus on our own wellness or wellbeing and that we’re taking care of ourselves.
They polled us recently, and loosely speaking, 50% of the people would like a more blended approach to work, so an office work from a home blend. 25% want to go back to the office, more the younger folk, because they’re young, and they want to go out at the weekends. 25% want a fully remote situation. That’s what we’re seeing at LinkedIn. I’ll share the Microsoft report as well. It’s pretty, pretty interesting.
We also did a remote survey similar to this one, and it was quite interesting that we saw responses were not unanimous. If you want remote, and some people explicitly don’t want remote, they’re not apathetic. They just want to go into the office.
If you go fully remote, you are going to make some people unhappy and exclude some people. I know that because I’m in the Netherlands, and the majority of my team is in South Africa.
It’s about workplace personalisation, so by giving people options now, same thing, it’s autonomy, right? We have to give that to people. I think that’s what our data showed, which was quite exciting.
Another question for me this time, Jonathan wants to know, is there a market trend of South African devs working remotely with foreign companies earning forex while staying in South Africa?
I do not have that data on hand. We do have the data of whether developers are interested in working for companies abroad.
But a lot of developers would take an opportunity if they were offered one, and I think one in three are actively looking. There’s a lot of people looking to move out of the country. We didn’t ask whether you would take a job in a different country. That said, we are operating in both Cape Town and Amsterdam, where I am, and we have companies that do hire South African Software Developers remotely.
That is a non-zero amount of people that are being hired remotely by Dutch companies and working in South Africa. I don’t have real train data on it, but it’s definitely happening, and at least for OfferZen in the Netherlands, we’re seeing that growth. Personally, sounds great living in Cape Town and earning euros.
I was just gonna say the same thing myself. If I could get a nice house in Cape Town on the sea and earn euros, perfect.
Yeah, that is some of the things we’ve seen. There’s at least the Dutch companies’ appetite for South African Software Developers, as it is in the same time zone, in English and culturally is very similar. There are a lot of factors at play, depending on how that all kind of unfolds, but I’m talking from opinion now, and we don’t like to do that.
Moving to the next question. Jeremy, what has been your experience trying to manage the work-life balance in the US with work and the kids?
It hasn’t been a problem for me because I’ve been extremely lucky to be privileged enough to work at companies that provide leave policies. For me, it’s never been a huge problem.
When I worked at Reddit, I had a terrible work-life balance, but that was by my own choosing and the fact that I was commuting two hours each way to get there.
Once I had children, I switched. I was working at Netflix first, and that definitely helped because it got rid of the commute, but I ended up spending most of that time just at work anyway. And then, once I had kids, I actually left Netflix to go work on my own, so that was how I managed it, but that was because I was in an extremely lucky position of being able to do that.
It’s hard around here, for sure. Most of the tech companies around here provide leave that’s very similar to Europe, so it’s not a huge problem if you are lucky enough to get a good job with a tech company around here, but if it’s not, then yeah, it’s gonna be a problem.
I’m interested. Did that work out how you thought it would because I know a lot of people, including myself, that when you have your first child thought,** “I’m going to work for myself, and I’ll have more time?” It kind of doesn’t work out that way often because when you work for yourself, you never stop.**
Yeah, I have more time for the kids, which has been wonderful. It’s problematic in a way, especially if you work from home where a lot of people, especially children, do not understand the difference between playing at home and working from home. So it’s really hard to turn down your three-year-old when he comes in and says, “Let’s play Lego’s now, daddy” and you well… I kind of do want to play Lego right now.
Okay, another question for Jeremy. What are some of the reasons or signals that made you turn down a job offer in your past?
Turned down a job offer? Have I turned down a job offer? I turned down one job offer, and that was only because I was choosing between that and Netflix and that other company was an up and coming startup, which is now a public company Cloudflare. It was basically choosing between being about number 20ish at Cloudflare or going to Netflix.
One of the main reasons I chose Netflix is just the commute. Cloudflare was in San Francisco, and Netflix was closer to home. Yeah, that was pretty much the main thing. I love the people. I knew the founders. I knew a couple of the engineers there. I knew it was a great place to work, so that was the only factor.
If I was going again, though, the main thing that I’m looking for is who I’m going to work for and what their management style is - are they going to be micromanagers; are they going to be kind of hands-off let me do my thing. Are they going to help me with career progression; are they going to help me find opportunities within the company, or are they senior within the company, and can they help me navigate the company? That kind of thing. That’s what I’m looking for if I go to a new corporate job.
I have a couple of ex-managers who I will definitely work for again. If they called me up and said I got a job, I would be there with no hesitation.
A question for each of you (lightning question - 30 seconds) what one piece of practical advice you would give someone who’s starting out their software career, in 30 seconds or less.
Find a project that lets you expand your knowledge in either systems or programming language. That would probably be the biggest thing. Look for opportunities to do something that you don’t know and aren’t comfortable with.
Work on your network: look for opportunities and work on your own personal network. It is so valuable, so that would be my one piece of advice.
Awesome, I think we’re gonna wrap it up there. There are a bunch of other questions in the chat. We’ll move them to Twitter. Find Ray, Jeremy, or myself on Twitter if you have any more questions.