📊 Recruiting Corner: Business Intelligence

All right. Well, hello from sunny Austin, Texas. I am so happy to be participating in the BI Takeover on Cloudflare TV and combining it with our bi-monthly show, The Recruiting Corner, to talk about all things business intelligence, data science and data engineering.

And so today I'm excited to introduce some of my colleagues that I've been fortunate enough to work with for many years as hiring leaders here at Cloudflare.

And so we're going to talk about the openings that they do currently have in Austin.

So I'm going to go ahead and kind of go around Robin so everyone can introduce themselves, starting with Nikole.

Hi, my name is Nikole Phillips and I am the data analytics manager here on the business intelligence team in our Austin office.

Awesome. Katrina. I am Katrina Riehl. I'm director of data science in the business intelligence team also in Austin, Texas.

And Chandra. I'm Chandra Raju.

I'm the director for data engineering. I'm also part of the business intelligence team based out of Austin.

Awesome. Awesome. Well, I think a great place to start off is, you know, how each of you ended up at Cloudflare.

So Katrina, tell us, since you were the first one to join out of this group.

All right. Absolutely.

I actually heard about Cloudflare several years ago, to tell you the truth.

I am part of a nonprofit organization outside of Cloudflare where we actually use Cloudflare on all of our websites.

And so as one of the more glamorous parts of my involvement there, I'm one of the web administrators.

So I've been receiving Cloudflare alerts for a really, really long time.

And so when the opportunity arose to come here, I was very excited to join Cloudflare and get some, you know, data science, you know, in the door.

So really excited to be here. Awesome.

And how long have you been here, Katrina? Almost two years. Two weeks from now, in fact, it'll be two years.

Coming up on that anniversary. Awesome. How about you, Nicole?

Okay. Yeah. For myself, it's interesting because I ended up learning about Cloudflare by a glass door alert and I seen a opportunity here.

So when I looked into what they're doing in the community, it made me even more intrigued.

So I definitely decided to apply and cybersecurity was one of the domains that I had not worked in throughout my career.

So that intrigued me even more because I like to be challenged and then being able to be on a new team and work with the colleagues that I work with.

It was a win -win across the board to be able to use data, to continue to let Cloudflare be a dominant force within the tech industry.

Awesome.

Awesome. And how long have you been here? It'll be two years, actually, in May.

Okay. Okay. I started a couple of weeks after Katrina. How about you, Chandra?

Yeah. For me, journey was also interesting. I've been in the travel industry and telecom industry before.

Secure domain seems very interesting with all companies now moving to cloud and security being one of the focus for a lot of companies.

And the amount of challenge they have with the new products and some of the scale of Internet they handle.

I want to see something challenging to build something new and interesting and scalable.

And how can we collaborate with partners?

The Cloudflare opportunity seemed to be very interesting. And the way they have been in everywhere, focusing on nonprofit, helping them to empower a lot of different organizations, this seems to be a natural fit for me to explore this opportunity.

I'm glad I'm here with a wonderful team of members here. Yeah, no, absolutely.

And I'll go ahead and let you all know that when I saw all you guys leave and go to Cloudflare, who is this company that's I've been working with.

And as I started to become more familiar with Cloudflare, I really could identify with the whole concept of building a better Internet.

The mission really was something that I had a passion for.

And security and working in the cloud are all things that are exciting and a transformation that just is happening globally.

And so when the opportunity to be the first recruiter in Austin came about, I jumped at the opportunity to work with you all again.

So great to put the band back together.

Yes. Let's talk a little bit about what you all do at Cloudflare.

So at a high level, Chandra, can you start off by telling us what the BI organization does?

Yeah. Thanks, Todd. BI's role is to unlock analytics capabilities and generate critical insights for our business teams.

How we work together as a team, cross-collaborate with different business partners within Cloudflare and allow them to be taking more data-driven decisions.

We collect data from first party and third party data sites, like how we convert the data to an information and insights, which can help our business to grow.

How we collaborate with them, giving them democratic data, give them tools to access that data in an easy way.

Our goal is to empower business to take that next level of data-driven decisions, which we have been doing successfully.

Within a short span of time, BI team has been formed.

As we said, we're just getting started. We have a lot to do, but we're making the right decision, right investments and collaborating on right use cases with our stakeholders.

I'm glad we have reached the stage where we can expand further and contribute further to the team success.

Awesome. Let's take a click down and talk specifically about the data engineering team Chandra and how that fits within the team.

I think BI has said we have data engineering, data analysts, and data scientists.

We all work as a team. Data engineering specifically, we focus on how we bring in different data sets from first party or third party data sets.

How the data can be brought in a way it's scalable, cleansed, and transformed in a way that partners can consume it, whether it's a data science team or a data analyst team or business partners.

The goal is to build that scalable pipeline, which is reproducible, repeated in automated fashion, and then work with our stakeholders to see how we can build also more Excel server tools to access the data in a more seamless way.

Our goal is to enable the data within the organization and then help others to use the data in a more effective way to drive further insights.

Awesome. And then from a data analytics perspective, Nicole, tell me a little bit more about your team.

Yeah, so once the data engineering team goes through their process, my team, what we do is we use that data from the perspective of engaging with the business to really be able to understand what business problems are you trying to solve, what business needs do you have, and then taking that information back and synthesizing it into a format where it can be consumed by our business users so they're enabled to be able to make data-informed decisions as it pertains to their business goals.

Now, it sounds simpler than what it really is because we have to go through a lot of EDA, which is exploratory data analysis, to really be able to understand if the data is going to meet the need of the business problem or the use case.

And that's where the collaboration with the big data engineers comes in and even and also our stakeholders because we have to all come to a consensus about the integrity of the data before the analysis even begins for us to be able to provide that insight to them.

Excellent, excellent.

And then Katrina, on the data science side, how does that tie into this? Sure, if you want to look at this as kind of a flow of information through the team, like data engineering brings all the data in, they make it usable, they collect a lot of data and history, things like that.

Then, you know, analytics comes in, they bring all the insights and make sure that people have the decision-making data that they need.

And then on the data science side, we really focus on automating decision-making.

So the idea that we can take all of that data and mine it and be able to create artificial intelligence -based systems that would allow us to be able to automate some of these decision -making processes.

So we really focus primarily on machine learning within my team.

Oh, okay. And so, you know, a lot of times in data scientists or data science, I hear about people being data scientists or machine learning engineers.

Can you tell me a little bit about the difference and how significant?

Sure. We actually, we have both on the team, by the way, right? So we did split this up into two different roles.

So we do have data scientists and their primary responsibility is going to be modeling.

So getting very close to the data, mining the data for really important patterns, understanding the data very, very in-depth and being able to tune and create the machine learning models.

Then on the machine learning engineer side, that's really the engineering piece that's going to unlock that, right?

So once we have a model, we want to be able to deploy it.

We want to make it usable. We want to integrate it with systems downstream, and we want to make sure that it's also integrated within our platform.

And so there's quite a bit of engineering in order to be able to make a successful project on both the data engineering side, as well as on the machine learning engineer side in order to make that entire flow work.

Okay, awesome. And I know that there are so many different tools and technologies that are common in the marketplace.

Katrina, can you tell us a little bit about the tools and technologies that we do utilize at Cloudflare and maybe just from the BI layer all the way to the backend?

Sure. So I actually, there's a laundry list of technologies, by the way, that we use because obviously this is a pretty big domain and you've talked, Chandra and Nicole and myself, right?

We actually have two major platforms that are under development within our organization.

So we have a data engineering platform, which can also be broken up into sort of our data ingestion framework, as well as our data access layer.

Then we also have a machine learning engineer, or I'm sorry, machine learning platform that allows us to model, to be able to deploy machine learning models.

And these two actually feed on each other.

If you think of it as like the machine learning platform as being both in a consumer, as well as a producer of the data that feeds back into our data engineering platform.

So within this whole stack, we actually have, like I said, just a ton of technologies that are being used.

And I know I'm going to skip some, but I'm going to try and hit the highlights as much as I possibly can.

But obviously we can't do anything without Google Cloud, right?

That's what kind of the basis upon which we're using everything. So we can, you know, we do have, you know, GCS as well as BigQuery on top of that, right?

So obviously SQL as well as Spark is going to be used in that.

The primary languages I would say that the team is using are going to be Python as well as Scala with that.

And so, and SQL, if you want to consider that a language, but that's a different story.

But anyway, like I said, we're going to be using Spark as well as there are some other tools around this that I also want to mention.

So have to talk about Kafka as being something that we're using.

We're using Docker, we're using Kubernetes, we're using Airflow.

Over on the data science side, we're also using Argo.

And we're also, we have RESTful interfaces and gRPC for remote calls back into the systems that are being created there.

And I also should mention on the machine learning side, we also are using Kubeflow, which is something that allows us to be able to ease some of this.

And then moving over to the data analytics side, for sure, obviously using quite a bit of BigQuery, also using Python, also using Tableau, Kibana dashboards, visualization through our own customized apps as well.

So we do have in-house visualizations and dashboards that we're creating as well.

And those are going to be web applications that react front ends, and then also having the backend using our existing infrastructure.

And with that one, I also want to mention that we're also using some Cloudflare products within our apps that are being deployed.

So just a shout out to Cloudflare Access, to Argo. Some of these things are definitely...

So we're using our own products within the applications that we're deploying.

And kind of the last piece I do want to talk about a little bit, since machine learning is kind of the domain that we're in, we're very much tied to the PyData ecosystem.

And so that's something that we spend quite a bit of time working in.

And that requires the scientific computing stack, which is going to include NumPy, Pandas, Scikit-learn.

We use a lot of light GBM and XGBoost, all of these other pieces that kind of go under the hood.

And we also have a tendency to move more towards PySpark as opposed to Spark and Scala that might possibly be used upstream from us.

And then one more language I did want to mention, since I'm trying to be as complete as possible, got to give a shout out to...

I know Go is also being used on the backend for some of the application, the web application work.

Yeah. I'd like to come and add to what Katrina said. It's a long list, she said.

If you look at the focus, majority of them, 90, 19% is open source technologies that we allow.

We try to pick a technology that is platform vendor agnostic.

So if I want to move from a cloud platform to an on-prem platform or hybrid platform, how can we move without having to have any vendor lock-in?

So that's a kind of goal design parameter we always put in.

How can we keep our designer tools and technologies in a very vendor agnostic format?

So I think that's what Katrina said.

It's kind of a long list of things what we adopt from open source.

Absolutely. There's a very great dependence as well as a support for open source within the team.

That's awesome. Awesome. Raise your hand if you're hiring. All right.

Looks like he's got multiple openings. Let's go ahead and start with Nicole.

Tell us a little bit about the core competencies and behaviors that are important for you to be successful on your team.

Yes. So I am looking for, at this time, early talent, which would be college graduates, people that are transitioning into analytics because I believe in the push me, pull you mentality is someone pushed me into this opportunity.

So I want to be able to provide that to someone else where they can apply their academic training within the workforce.

And then I'm also looking for mid-level candidates, which is someone that is about three years within the analytics domain.

If you're looking, please go on the Klaffner Jobs website and apply if you're interested in our roles.

Now, what am I looking for in a data analyst?

I definitely need someone that enjoys solving problems. I need you to be a critical thinker where you're trying to put yourself in the business user's mind and not only address the use case or the problem, but think about what other answers that you could provide to them, like those what ifs.

Those what ifs will always turn into more what ifs so we can continue to enable our business to make data-informed decisions.

I also need someone that likes to tell a story with the data, someone that can tell the narrative and be able to flow, but then also be able to iterate on that story as things change because the industry is going to change, the domain is going to change, so you're able to iterate your story as that happens.

From a technical perspective, as it was stated, we do use Google Cloud, so BigQuery is one of the primary skills that we need.

And I'll say SQL.

I'm not necessarily married to the tool, but more so the skill set because you can get ramped up on the tool if you have the fundamental SQL skills.

And with the large volumes of data that we have, there's no way in any possible way that we can do analysis in Google Sheets or Excel, so those are skills that we definitely need.

As it pertains to visualization, I definitely need data visualization skills. We do use Tableau as our enterprise-wide tool, but again, I'm more concerned about the skill set than the actual tool because you can get ramped up on the skill.

In addition to that, Python or R, preferably Python because we are a Python shop because as we continue to mature as a team, we will be partnering a lot more with our data scientists and having that skill set would make that partnership go smoother.

And lastly, just as it pertains to anything from the soft skills perspective, I want someone that brings something different to the team.

Of course you want to have the fundamental and the core skills, but my current data analyst team, they are dynamic and they are fantastic, but they all have something unique about them that makes our team really gel.

It's kind of like a cake. You can't use the same ingredient or it'll be nasty, but if you have different ingredients, that's what makes the cake delicious.

And that's how I look at the data analyst team.

I love that metaphor. And if I were to give anybody advice who is applying for data analyst roles, I think one of the shortcomings that people sometimes miss on their resume is they just kind of list the activities that they did, how those activities ended up showcasing results.

And so I think for data analysts, you want to be able to kind of tie all that in there and what that impact to the business, as opposed to just a laundry list of activities that you did.

Yep.

Thank you, Todd, for that. Yes, that's very true. You bet. Chandra, tell us a little bit about your opening.

Yeah, we've been always hiring. Todd, you're the one that's always with all this back and forth with hiring.

We're hiring a mid -level data engineer as well as a senior level data engineer.

So based out of Austin, it requires strong data engineering skills using Spark, Scala.

So that's one of the key skills that we look for.

And then SQL is always there. That's kind of a DNA of any candidate that we look for, strong SQL skills.

And also the skill technique, always we like to think about resource when we look for is how can they marry technical skills with their interpersonal skills and then business development skills.

That's a focus we always do is how they can combine all three categories and bring in that kind of new value to the team, how they can bring in the new talent or capability to the team that the current team might need to develop further.

So those are the candidates we look for. And then somebody who has done this in a scale, because a lot of scenes probably we see like, no, they've done this, but have they done it in scale, in a production level grid, looking at from end to end, from perspective, starting from collecting data until business gets impacted and what kind of business impact they're able to make, how they collaborate with different stakeholders, how they even communicate their ideas to the team.

So that's big. We can do a great work if we want to communicate the ideas.

So we look for those talents who can share the ideas and communicate and present to the executive team or broader audience.

So those are different talents we do.

And then also we have a balance between teams. Like we have a lot of senior engineers and new grads and new engineers join our team, how they can be an effective mentor, because that's what we try to put balances like being an individual contractor is great, but are you being a great mentor, a team player who can work with multiple teams?

Because as Nicole and Tetsuna are saying, we work as a team, how we can work across, collaborate within the team and bring that value and then take that to go as a one friend when you're representing the business and the stakeholder, how that can be established.

So it's always a balance we look for, like technical interpersonal skills, soft skills, and then business enablement skills.

How can they quickly learn a domain and enable them to be a successful engineer?

Awesome. Awesome. Katrina, tell us about your openings. Yeah, absolutely.

So we are hiring currently for both machine learning engineers, as well as data scientists.

We are looking for more mid-level people and some entry-level people with great potential.

So that's the area that we're really trying to focus on right now.

I want to say all of that, that was mentioned, right?

We want all of those things, but we also really within the data science team, of course, we want a really strong background and fundamental knowledge in data mining, machine learning, and also visualization techniques.

What we're really trying to do is to democratize our data and make this less scary.

The idea that we have these automated systems on the backend that are able to produce insights and to help with decision making and to automate decision making and things like that, really work with our stakeholders to make sure that we are providing that business value, but then also making sure that it's usable and that it's also very accessible, right?

We want people to interact with the data and get very, very deep within it.

So that also leads into the communication skills that were really alluded to by both Nicole and Chandra here, right?

That on the data science team, it's very important for us to be able to not only quickly learn a domain, right?

But also be able to communicate with our stakeholders, make sure that we can speak in that business value way, but then also be able to turn around and be able to work with our engineering staff and be able to have that engineering background and vocabulary in order to be able to actually actualize these systems, right?

Because we're really trying to unlock this technology in order to make it more accessible to the people who need this data and need these insights in order to move forward with their jobs.

That's awesome. Nicole, maybe you can share an example of a challenge that the team has either already solved or is currently solving that would be exciting for people that are looking for new opportunities.

How much? We have eight minutes? Okay. So the first thing I'll say is the ultimate goal of the challenges we're trying to address is we want to be able to provide self-serve for our business users.

So some of the things that we are collectively working through as a team is really being able to understand the behaviors and patterns of our customers and as they journey through Cloudflare because we want to be able to just as an example, the marketing team with who they target for their marketing campaigns and they're not just generalizing a target or their sales team as they're engaging with their customers.

They can have a more customized discussion because they have the data and insights at their fingertips with just to say our infrastructure team helping them understand how traffic is changing over our network so they can do server capacity planning in a more data-informed way because I always say the infrastructure team is like the nucleus of Cloudflare.

Without them, we are done. There is no Cloudflare.

So we really need to be able to have our network up and running optimally.

And then also helping our executives understand the health of the company but also working through what success looks like and what those metrics are and how that may change over time.

So those are just some of the a few things that we're working on and it's just the beginning and we're going to continue to iterate on that.

That's awesome. That's awesome. Chandra, tell me about something that you've learned during your time at Cloudflare that you think will carry with you the rest of your career.

Yeah, that's a kind of there's a lot of it's very interesting journey.

Everybody says a lot of learnings from Cloudflare. First, in level of scale that Cloudflare handles, the level of products and new services that we deliver in a space unimaginable.

A lot of people have been saying Cloudflare is a pure engineering company.

We can really feel it, how Cloudflare being that kind of culture, how quickly they get an idea and deliver a product and which is enabled for a wide variety of customers across all machines, across geos.

That gives us a great motivation to see how Cloudflare has been fast paced and doing that in a scale and with such a vigor.

That's kind of what we feel like, okay, how we can do the same thing.

We have a unique opportunity to cross collaborate with so many business partners.

We have a unique opportunity kind of doing it at scale because a lot of data-driven enabling capability we can enable for our customers.

We have a unique opportunity to kind of collaborate at that level. This opportunity kind of gives that scale where we can do with different use cases, whatever things we want to enable our partners to reach that level.

That is kind of giving it more, with the amount of data and level of sophistication we do with data science models and critical insights that Nico's team is doing, how that can be enabled, making a difference, giving that from ground up, that's what we're doing.

BI platform was built from ground up.

That's what the previous sessions we did. All were done and how we were able to do in a short time, it's all collaboration with different business partners that we could get happen.

And because the culture of Cloudflare is always collaborative and transparent, which is great, helping us to further expedite that enablement that we're able to give the business, which is a great experience for me to carry forward and which will be like continue to grow as I spend more time in Cloudflare.

That's awesome. That's awesome. Katrina, how about you?

Something that you hadn't had the exposure to or done before in your career that you know that you'll carry with the rest of your career?

Yeah, absolutely.

I do want to reiterate what Chandra just said. The scale is unbelievable here.

Just the sheer volume of data that Cloudflare produces. And it's really quite humbling to tell you the truth.

And so that's for sure something that I will for sure take that forward in my career.

I also just want to mention that there is a great opportunity, I think, within data sciences may be a little bit newer and a little bit more foreign to people that we're working with.

So that opportunity to educate, to be able to work with people, to show capabilities, to be able to really provide some thought leadership and how we might want to move into machine learning and artificial intelligence moving forward has definitely been an incredible opportunity for me.

But also I think it's taught us a lot about going back to what we were saying about making this accessible and maybe making it a little less scary for people to understand this process a little bit better.

I think that that's for sure something that I'll take with me as I move forward.

That's awesome. Nicole? Yeah, I'll keep it short. What Katrina and both Chandra said, but also the fact that we're working through ambiguity.

So I would say keeping it simple and making sure you're not just looking for the new and shiny technology, but keeping it simple as it applies to the need so that we can push out what the product is for our end users.

And that's something that I'll take with me just in general is don't look at the new and shiny thing, but look at what is practical and what is simple to create the solution.

That's awesome. That's awesome. And in terms of just getting to know each of you, a little bit more on a personal level, when you're not working, what are things that you do to keep yourself busy and entertain?

Nicole, why don't you start? I have a pretty simple life. My family is my everything.

So number one, taking care of them. Other than that, I love reading.

So definitely have my reading list for the year and enjoying my back patio before it gets too hot outside in the Texas summers.

I got into the home gym craze when the pandemic hit.

So that's one of them, but that's pretty much it for me.

Awesome. And when the pandemic does end and you're safe to travel again, where are you excited to go?

The one place I do want to go back to is Denmark because I never got to enjoy it the way I wanted to when my husband and I got married because I was so cold.

So I would like to go back when I'm actually prepared for that level of cold.

For London, Mark. Yeah. Awesome. Chandra, 30 seconds. Same question. Yeah. For me, it's more often I just started biking.

So last summer I had a success in teaching my daughter to learn biking.

So this summer I signed up for teaching my son to ride the bike.

I know he's four years old, but I know it's going to take him some time, but I hope I'll make it quicker for him because that helps me to bike with him.

Otherwise I've been walking with him. So I want to start biking with him together.

Awesome. Katrina, bring it. Sure. I actually spend quite a bit of time with nonprofit organizations doing data science in my spare time to tell you the truth.

So I'm involved. I talked a little bit about numb focus earlier, at least alluded to them, but that's a scientific software open source community.

I also work with being able to help with some of the gerrymandering issues that we have here as well as helping out with human trafficking problems that we face all around the world.

And then as far as beyond that, I'm also a aspiring writer. We're out of time and we may or may not go live, but thank you so much.

And if you're watching, reach out to any of us on LinkedIn if any of these opportunities resonate with you.

Thanks so much for joining. Thank you, Todd. Thank you. Have a good one.

Bye-bye.