Cloudflare TV

📺 CFTV Anniversary: Cloudflare Careers Day - Engineering & Product at Cloudflare

Presented by Aly Cabral, Joaquin Madruga, Tom Lianza
Originally aired on 

Aly Cabral (Director of Product Management, Workers) Joaquin Madruga (Engineering Director) and Tom Lianza (Engineering Director) will represent the Product and Engineering Team to discuss their hiring philosophy, share profiles of great candidates, and give advice on building a career in Product or Engineering.

Join our LinkedIn Event: http://cfl.re/CloudflareCareersDayNAMER

To find all of our current open roles, go to https://www.cloudflare.com/careers/jobs/

English
CFTV Anniversary
NAMER Careers Day

Transcript (Beta)

Hello, and welcome to our career day session on the product and engineering teams.

So the first thing we're going to do is go around and introduce ourselves and talk about what originally brought us to Cloudflare.

So first things first, I'm Aly Cabral.

I am the director of product for our workers platform. And prior to working at Cloudflare, I worked in the database space, and I was looking to kind of move upstack and have like more of an impact on the developer lifecycle and how developers solve their own problems.

So very excited to kind of join specifically the workers platform and enable developers to solve problems.

Really at the end of the day, that's what we're all about.

Joaquin Madruga Great. I'm Joaquin Madruga.

I'm actually Aly's counterpart on the engineering side. So I'm a director of engineering for workers.

I've worked on a few different projects here. And actually, when I first came to Cloudflare, it wasn't for workers.

I joined a little bit later in my tenure but a lot of what was exciting about Cloudflare is just kind of the scope and the breadth of Cloudflare.

Even when I joined, I guess about three years ago, it was still backing a large portion of the web and we're backing a lot more of it now.

And it's just really neat, especially with the workers platform to see not only how do you accelerate and make sites quicker, but also how do people build apps and deploy that on the web at scale.

And it's been a lot of fun. Tom Lanz I'm Tom Lanz.

I'm a peer of Joaquin's. I've been here just about six years.

And when I started, I managed the SSL team. So universal SSL, the free certificates that we gave the world.

And before that, I was a customer. So that's how I got hooked.

And joining for me was really similar to Joaquin. I guess the impact was astounding.

It continues to amaze me how few people work here relative to how large an impact we have on the Internet.

And it's very hard not to have an impact in Cloudflare.

There's just no room not to. We all work whatever at another company you might think is a team, at Cloudflare is a person and so on and so forth.

And it has been nonstop. We've gotten bigger on the Internet faster than I think we've gotten bigger as a group of human beings, which keeps it exciting.

And certainly kept me very engaged over the years.

Yeah. And one of the exciting things is like, not only do we have a large impact now, but the growth is so exponential even still, right?

It's like us doing our jobs, right? Our new users are going to greatly outnumber our current users, which is a massive user base of today.

So it's all just like a compounding impact over time as well, as we also develop and ship new products too, which is pretty exciting and service our current customers as well.

Now, one thing we didn't mention, and the first question there was our location.

So maybe going back around with where we're based out of, and then if we have team members in other locations, that could be a good thing to touch on.

So I'm based out of the Austin office along with Joaquin, but I have product managers on my team and developer advocates that span a couple of different cities.

We have someone in Vancouver, in New York City, in San Francisco, and someone in Melbourne even.

So a widely expansive team, but most of the folks are co-located in Austin for us.

Yeah, same on the engineering side for workers as well. One of the interesting things kind of about at least engineering at Cloudflare is if you kind of watch how a request goes across our network, it's almost the opposite direction of actually where people work at Cloudflare that work on it.

So we've got a large London office and a lot of like the initial requests when you connect to Cloudflare, a lot of the people that actually work on that are there.

When it comes into workers themselves, we're really based in Austin.

So a lot of the stuff that we do and build out there is really centered in a large part around Austin, but we do have a lot of people in other locales, and then a few people out in San Francisco as well.

And then we have our fair share of remote employees now too, especially those that have onboarded here during the pandemic.

And I'm based in our San Francisco office, not that I've seen it in quite some time.

And my direct reports are largely in North America, but the greater team is everywhere.

I oversee a lot of the control plane, the API, and the back ends for where you store the data that Cloudflare, where you tell Cloudflare what to do.

And there's a large operational component to that.

So we have our API team, for example, has people across all the time zones in the United States, also in Australia, also in London.

And it has the benefits of being able to follow the sun and make sure you're not waiting.

You can sort of cover for each other while people are asleep. Downsides are you only all get together once a week at a time that's mildly inconvenient for everybody, but nets out.

And it's certainly something we have to get better at as a company.

I wonder how many of the time zones we have humans in at this point, but we are definitely getting more and more coverage around the world across our teams.

Awesome. So how long have you been at Cloudflare, Tom? And has your role changed or developed in that time?

Yeah. So for me, it's six years, started as a manager for SSL.

When I joined, Cloudflare tries to stay, I think as flat as possible.

We're not a company with a lot of flashy titles. We're not very sort of deep and leveled and sort of try and keep that startup ethos focused on the work.

We eventually got too big where just structurally we needed to create a level.

We couldn't have the head of engineering with 20 director ports, just like anyone else, and created the director level.

And a few of us moved into that space and we fleshed out the org that way.

So now I oversee a bunch of the SSL team, still part of my life, but also database, Kubernetes, and some of the core platform for the control plane API.

How do you see the EM job differing from the director job at Cloudflare?

When I joined, the thing I felt was always missing when we were just a bunch of EMs, we were all focused on shipping, getting our stuff done and out.

And I felt like there was a gap in there wasn't anyone whose job it was to think about how we are working.

It was kind of like a best effort. How should we be doing recruiting or how should we be working together?

We were working on the doing and there wasn't sort of room to think about the how.

And I think that's where directors, Joaquin can back me up or contradict me, but I think we do try to work on the how.

We're trying to be supportive and enable those teams to continue shipping and let them sort of offload some of the, like, how is this supposed to work?

This is frustrating. This is painful. Help me with this, that, and the other thing, where when we were a smaller company, you just kind of rolled with the punches, I felt like.

Yeah, I agree. I mean, a large part as well is if you have a bunch of individual teams, they're, you kind of, you fixate on what you're responsible for.

And it's really difficult to figure out how do you launch new things or how do you absorb things that have been there for a while.

So we spend a lot of time figuring out what's the best way to structure, you know, really the organization around the problems that we want to solve.

And one of the nice things is with a lot of growth, you know, that we see, I think we've doubled in a year, right, in terms of head count.

And so there are always new opportunities for, to work on some things.

And it means that you're usually when we split and kind of organize around things, we're actually splitting teams and trying to focus on different areas.

And so somebody needs to be mindful of that and think through that and think of the impacts.

And we do spend a lot of time on that.

Yeah. And speaking of like new opportunities and new challenges that people can take on throughout their time at Cloudflare, let's touch on some projects that the teams are currently working on.

If anyone follows along with Cloudflare, they know that we ship a lot of product.

And that's a cultural thing that we really do focus on refining and focus on that innovation culture within Cloudflare, really on every team, which is a really fun thing.

And on Workers specifically, we've built this highly distributed compute platform.

That's really, really powerful, right?

When you deploy a worker script for the first time, you're deploying in all of Cloudflare's 200 plus data centers automatically without needing to do any configuration at all.

You're just online at global scale, just with a single deploy.

Now, one of the interesting challenges that we've really focused on since really nailing that compute piece is how we deliver state to that compute.

And like, how do we get data accessible on that edge so that you can start to do very complicated full stack applications directly on Cloudflare's infrastructure?

And we started by shipping a product called Workers KV, which is our distributed global key value store.

We've added Sense Durable Objects, which is currently in open beta, and it's strongly consistent state on the edge and introduces new use cases around coordination and other things that you need for that strong consistency.

But those are only two data options. And in order to kind of satisfy the full suite of applications we want to satisfy, we'll need to build out a lot more.

So we're starting to kind of like nail down from a roadmapping perspective and strategic perspective, what comes next in that data space, and then rallying a team or folks around those initiatives to kind of kick off new products in that space.

And that's a really exciting thing that we're focused on on the platform.

Yeah, I agree. It's an interesting kind of role to hire for too, right?

Because it's a mix. It's somebody that wants to get their hands dirty kind of by necessity, right?

When you have a team of two or three, just one person just solely managing that doesn't make a lot of sense.

But also somebody that understands that as things grow and as we launch, things get big here really quick, right?

We have a lot of users.

We have a lot of people that are interested in pretty quickly.

We've actually spent a lot of time trying to engineer how to measure how we roll out products.

And so as that goes, we want somebody that can actually try and grow the product at the same pace that it's being adopted.

And that can be an interesting test.

It's very rewarding, very fun. I've been able to do it a few times here, and it's neat to try and find somebody else to be able to help run some of this stuff as well.

Yeah, I feel like you guys are inventing the future, and my team's job is to make sure you are enabled to do that.

So a lot of my teams use technologies that everyone's familiar with.

Postgres, Kubernetes, API gateways, NGINX, load balancers, the console, all kinds of fun, open source infrastructure.

And we try to create a solid, stable foundation using known technologies so that the rest of Cloudflare can invent the future.

And so a lot of the projects we're working on involve piecing those technologies together to ensure we have an extremely reliable control plane to a global...

We still have the problem of state.

I don't think anybody, any of us has yet solved all of the problems of state.

But my team store the state for where all your DNS records that you want us to have, those sorts of things.

And so using primitives like console and DNS and database and Postgres to allow teams to build products in a way that's stable and highly available are what my teams do.

And that's where a lot of our projects are focused is on reliability.

And to be honest, compliance, a lot of compliance. Our customers want us to check a lot of boxes, make sure we were audited up and down for every long list of alphabet super compliance.

And so we work on making sure we can meet those expectations as well.

That makes a lot of sense. One other project that we're working on in the workers space, though it's not directly under the workers brand is Cloudflare Pages.

We launched in April Cloudflare Pages is our opinionated way to build and collaborate on and deploy websites just from the ground up, starting with static sites.

And you can add a worker to kind of add that dynamicism into your website.

And we're super excited about the adoption. Pages went from no users to thousands of users overnight.

And seeing that as such a kind of fulfillment feedback loop, right?

Like that's quite, it's funny because I'm coming from companies where we couldn't get anyone to care about new features we were building for a long time.

And to see that impact and that spike in traffic or spike in adoption immediately after launch is such a satisfying thing that we can take for granted knowing that like this is quite a unique environment that really we have a very captive user base that wants to see these new features and wants to adopt them as soon as we have them.

And so very excited about Pages, but we're just getting started in providing those opinionated experiences on top of our foundational technology.

We want to provide opinionated experience to work with ML, right?

Like in the future and opinionated ways to build a mobile app, right?

Like all of that very future looking, but you could see us moving in that direction as well with the success of something like Pages.

So very excited about what the future has in mind there too.

I never told you this, but one night I woke up in the middle of the night.

I just couldn't sleep. I built on the weekend, this static site.

It's static, but it calls an API to get the tides of my neighborhood Creek.

And it was just sitting on my, I just had it checked into GitHub and literally in bed, I was looking at my phone.

I was like, that's right. Pages lets you publish.

I published that site to my own domain on the Internet at three in the morning from my phone in bed.

And it just worked. It was unbelievable. Like I felt like you guys created magic and it, it's totally works to this day.

I woke up the next morning.

It was like, did I dream that? And it was there, there was live.

It's on my own sub domain. It was, it was amazing. Oh, I'm really great. That's really great to hear Tom.

Yeah. And that's the first time I've heard that story.

So that's, that's awesome. Pivoting a little bit, what do you guys look for in an ideal team member on, on y'all's team?

And I know that you have different teams that do different things.

So feel free to take that question as, as you will. Yeah, I think want of, of impact is, is a big one.

And, you know, we, we, we all talked about it.

It was a, it was a big driver for, for bringing us here, but somebody, you know, that really wants to change things and build something big.

It is, is, is certainly a, you know, a big driver, even more so than a particular language preference or any of that.

You know, we, our, our tech stack is pretty uniform because we tend to pick the right types of tools to build the right things that we're doing here, but people can learn that.

But really finding somebody that really wants to own a project and run it all the way through and see it launch and find other Toms at 3am that want to put their, you know, their Creek software online is, you know, it's a big driver.

In certainly when I'm, I'm talking to people and screening resumes, similar to what Joaquin said around ownership, but that means all of it.

That means have you been, have you shipped something that went down and failed and you do really, I got to get this thing back up.

Like you really do feel you weren't building a thing and then handing off for someone else to sort of own on your behalf or operate on your behalf.

But like, you really do feel pride in what you build and you want to see it all the way through.

That's, that's really important to me, I think.

And, and I do ask for stories like that. It can be anything too.

It can be, you know, a college club and you ran the website and, you know, you had Pingdom alert you that it went down or something like that.

All, all good.

It's just, you know, that passion, that care that comes with ownership is something that's important to me.

Yeah. Yeah. And for product managers, we look for things like technical curiosity, like making sure that you're willing to get your hands dirty, go into the weeds, use the products, of course, that we're shipping make sure that you can empathize with users, especially on our team.

Our users are the helpers. So getting that user empathy and having that technical curiosity is very important.

And then having that customer facing experience is also something we look for on the product side.

We are serving other companies most often here at Cloudflare.

So we have these high touch relationships with those companies and making sure that someone who is on my team is very comfortable talking and owning those escalations if they happen or customer relationships as we build out a roadmap or work with someone on validating a project.

That's incredibly important for my team as well.

We also hire developer advocates. And what we're looking for in that space is someone who's super passionate and maybe has built well, often someone who's built something on workers before, maybe open source some projects, very competent in writing, tutorials, writing blog posts, all of that good stuff is what we look for on that team.

Yeah. I'd say very technical as well.

Our engineers are obviously technical, but the engineering managers and product side are as well.

It's very common for product people to come from an engineering background.

And I think that helps. It helps with a lot of speed as well in terms of what we're trying to tackle and understanding how we build things too.

If you're an engineer, it's a great place. Some engineers will come from a place where they feel like their boss or the product people aren't engineering enough, but our product is other people's infrastructure.

There's no way around it being technical.

And you will find strong technical people up and down the organization in places you wouldn't normally see it at other companies, which is great for engineers, I think.

Yeah. Yeah. It helps you with the translation. Right.

So I think that's good. What open roles do you guys have right now? So we're...

Workers is growing fast, right? And so we're... In the same way that I described where we're splitting these teams, we're looking for engineering managers in order to be able to help lead some of these new projects.

That storage thing that Allie just teased is certainly one of those.

And we have a few other openings as well, just around...

A lot of the stuff around dev tooling and just how do we make the workers platform a little bit easier to use.

And then we have engineers all through...

Most REST APIs at Cloudflare are served by some sort of Go service somewhere.

A lot of the stuff that ends up handling packets tends to be REST or C++ developers in there.

We do abuse Nginx quite a bit here too. But really, that's the core set of languages on the backend side of things.

And then obviously, JavaScript and TypeScript, it's the web.

We do a lot of that as well. But really, any developers that are interested in any of those getting into that tech or have that tech background, we almost always have a place for them depending on what they're interested in on the project side.

Yeah. For some of the roles that we're looking for, people with Kubernetes experience for sure.

People with platform experience certainly...

I've seen a few resumes where somebody's like, I actually have a Raspberry Pi cluster running in my bedroom.

You're probably... You should come talk to us.

If that's what your curiosity is and your passion and assembling open source infrastructure, even if it's just Raspberry Pis, you'll have a blast here, I think.

So CI, CD, the security around all that, all the security, securing the infrastructure and the pipeline of a software development chain, all of that is...

We have a lot of need for more help there. And it's never been more important, as you can probably tell by reading the news, to make sure we are leaders in that space and are hiring people who are passionate about doing excellent work in infrastructure and infrastructure security.

Yeah. What about you, Allie?

Great. Yeah. So I have a product management role that's really focused on making our compute platform the most powerful compute platform that exists.

That means extending script limits. That means upping memory limits, just getting our platform to run any type of workload you may want to throw at it.

Also things like adding, for example, maybe one-day queues on top of the platform, things like that.

Owning those projects end-to-end, making a very powerful ecosystem to run your application and compute.

Definitely, if anyone's interested in taking ownership of that strategy or has experience working on developer-facing products before, would love to hear from you on that.

And then we are hiring developer advocates.

So folks who are very comfortable working in a community setting, working with a ton of users, we often on the workers team will engage via our Discord channel with many users.

At this point, it's like 4,500 users.

We're going to do a big celebration when we hit 5,000. And we just started earlier this year, this Discord channel.

So very excited about the growth there. We want people who are going to jump in, engage with our user base, nurture that excitement around our platform, and then contribute content as well.

So making sure that we're contributing docs or tutorials, and also adding to the Cloudflare blog, which is massively red.

So very excited about those roles on the team, for sure. Yeah.

In fact, with the community, a lot of times when people are interested in Cloudflare, specifically workers, I'll direct them onto Discord.

And you get a pretty good feel of how things work.

A lot of our developers hang out there. It's a great way to get a feel for how customers use it.

And it's also a good way to see how we release features and preview them.

And you can get some beta invites if you hang out there as well, which is always pretty fun.

Yeah. It's a good way to meet us, for sure.

Anyone can join. And very excited to see, especially if you're building applications on top of the platform, we'd love to hear from you and have you guys ask any questions there as well.

Is that a good way for people to get your attention from a hiring perspective as well?

It might be a little too much noise for me to keep up, if I'm being honest.

So maybe not, but it is a good idea to get a sense of the culture and see from your perspective whether this is something you're excited about too.

It's more of an opportunity for you as the candidate to get a good sense of us, the team.

But that is a good point. When people do apply, there's usually a pretty high correlation with people that are really excited about Cloudflare and use our products and people that work out really well here.

And so if you are putting in, whether it's in the cover letter or even just on your resume directly, the things that you've done with Cloudflare, the fact that you've used us before is always a really good way to just for us to be able to kind of understand what you're looking for and what you're excited about.

Yeah.

Are there any other good tips for standing out? I definitely think, I know a lot of people we talked to have read the blog.

I also worry a lot of people we don't talk to have read the blog and decided it's so scary because the people here, you have to write flawless kernel code to find a home at Cloudflare.

And that's not the case. Hopefully people heard that from our different teams.

We have a lot of different things we do here. Full stack, front end, infrastructure, brand new stuff, and all flavors of languages that don't let the fact that we have great experts deter you from applying.

If so, you should be enthusiastic about the fact that you can learn anything once you come here.

You can talk to world-leading experts in any one field. That's not the bar to be hired.

That's just a perk of joining. I think focusing on what you've built as well, and a little bit of what Tom had said is too, around what have you owned and what if you weren't there, wouldn't work, would fall apart, or that you've automated so that you can work on something else is always really helpful.

And as much as you can provide to direct link in there.

We're not big on looking for certifications or any of that kind of stuff.

It's really about owning and building product.

And so, anything that you can do to try and highlight that, whether it's a nice webpage that shows it off or describing some of the stuff in your cover letter, just any ways to kind of bring that up to the top has been really helpful.

We'll always ask a product manager who's in our pipeline or loop to use the product and come with feedback on the product.

So, one of the things that you can't ever get back once you've been at Cloudflare long enough is your fresh eyes, right?

Your new eyes, using the product the first time, feeling the pains that your new users feel, and getting that empathy for them.

And you can learn a lot about a candidate, too, by seeing what they kind of touch on and how they articulate their empathy for the user at the end of the day.

So, using the product, whether that's workers or pages, and getting a sense for that user journey on our platform is something that I would highly recommend for any product candidate.

We're running about out of time, so it's probably worth just describing how to find our teams.

And if you're interested in applying, if you go to the careers page on Cloudflare and search for workers, you'll be able to find anything that Ali and I work on.

Tom, what's a good way to find some of the posts that you're looking for? Yeah, if you look for Kubernetes, you look for security.

Kubernetes, you'll find just our teams.

But security, API, really, some of our posts are generic intentionally, just to get you in the funnel, let us sort of help you find the right roles so you don't have to do it yourself.

So, I definitely encourage you to find something that looks the most right to you and let the hiring manager help sort of steer you to the right home based on your interests.

I think we do a pretty good job with the phone screens and resume screens.

And if you write a good cover letter, that'll help us find the right role for you, even if you might otherwise be intimidated by the volume of hiring we're doing.

We are certainly growing fast. So, I understand the challenge of that wall of openings.

That's a good point, too. When you applied, if you're interested in one thing in particular and in talking to a hiring manager, let them know.

But also keep an open mind and be flexible there, too. We've got a lot of cool projects that may or may not be public or any of that and may be a good opportunity for you as well.

Absolutely. Great. Well, awesome. Thank you all for tuning in.

We're at time now and very, very excited to watch you guys all apply.

All right. Thanks, Tom. Thanks, Joaquin. Great talking to you guys. Thanks, Sally.

Thanks, Tom.