Top DevOps skills and technologies (and how to learn them)

In this post, we explore the soft skills and technical skills needed to succeed in DevOps careers — and how engineers both current and aspiring can work to develop them.

Employers want people with DevOps skills, and people want to learn DevOps skills. (“DevOps” is one the topics people searched for most on ACG in 2021.)

But current engineers and aspiring entrants to the field should know that DevOps is about more than just the latest and greatest in technical knowledge and ability. Success in a DevOps role requires a mix of soft skills and technical skills.

Accelerate your career

A Cloud Guru makes it easy (and awesome) to level up your career — whether you're a novice or a certified guru. Check out ACG’s current free courses or get started now with a free trial.

There are several technical skills essential for succeeding in a DevOps role. But DevOps job roles also require strong communication skills, adaptability, and a willingness to participate in activities like hiring. So what does it really mean to be a DevOps engineer, and how can someone acquire the skills needed to succeed in DevOps careers?

Let’s dig in!

DevOps, at its core, has one critical goal: enable faster, more efficient, and more secure software delivery and iteration. It does this via the marriage of “Development” and “Operations”. (Note: DevOps is not the same as agile.)

Legacy methods of software development involved long cycles of work before release, with software developers working in isolation from the system administrators and operations staff that would need to deploy and run it. Releases were often error prone, and feedback about application performance and compatibility took a slow and meandering path back to development teams.

Want to learn more? Check out ACG’s DevOps Learning Path to get started with DevOps basics or to level up your existing DevOps IQ.

DevOps engineers bring a unique combination of skills and experience to the table, often combining system administration, operations, and software development into a potent array of capabilities.

DevOps engineers typically work closely with development teams throughout the software development lifecycle (SDLC) of a given application or service. Automating build systems, deploying and running CI/CD infrastructure, and monitoring critical production applications are just some of the jobs they are tasked with.

While technical skills are important to DevOps, they don’t tell the whole story. In an environment where release cycles are compressed and new features involve multiple teams working together, a strong culture — built around people — is critical.

The answer is a resounding “yes!”

DevOps work can demand top-notch problem-solving and communication skills, but it remains one of the most fulfilling and rewarding jobs in modern software engineering.

Glassdoor reports that DevOps is one of the top 10 top-paying jobs in IT — and they also call it one of the “best jobs in America” based on salary, job satisfaction, and number of job openings.

This one also lands firmly in the “yes” column.

Putting the debate about using “DevOps” as a job title aside, it’s clear that the demand for engineers with this skill set has increased at a blistering pace. Marc Andreessen famously said “software is eating the world,” so it’s no surprise that a skill set aimed at driving faster adoption and release of software products and services garners a lot of attention.

• Randstad’s 2021 Salary Guide reports that demand for DevOps skills are skyrocketing, with 50,000 jobs posted for DevOps developers in the 12 months before their report, and an average salary of $137,830 USD — beating out even high-paying cloud engineer jobs.
• HR consulting firm Robert Half Talent Solutions lists DevOps Engineer as the second highest-paying IT job.
• Indeed reports the average salary of a DevOps Engineer to be around $118,000 USD as of January 2022.

The opportunity to work with cutting-edge software technology, as well as the allure of “top-of-market” compensation provides an equally strong draw for job-seekers.

Get the Cloud Dictionary of Pain
Speaking cloud doesn’t have to be hard. We analyzed millions of responses to ID the top terms and concepts that trip students up. Gran this cloud guide for succinct definitions of the most painful cloud terms.

Soft skills are essential for success in DevOps careers. But they don't tell the whole story though.

An engineering team can be a well-oiled machine, dialed in on culture and robust communicators, but if they're using legacy tools and software, they're not going to get very far in building modern software applications. Current and aspiring engineers need to focus on developing modern, in-demand technical skills as well as their soft skills.

There's a saying about DevOps that it's not a tool, or even a job description. The main thesis behind that principle is to reinforce the idea that DevOps is about working holistically towards the goal of faster, better software delivery. Reducing it to a single tool, or a company saying they "do" DevOps because they hired a couple of "DevOps" engineers misses the point, and generally ensures none of the ultimate goals are actually achieved.

Still, it can be useful to think of DevOps as a common set of tools, software, and design patterns.

Compared to more legacy paradigms of software engineering, DevOps is a relatively newer entrant to the scene, having been introduced in 2009. DevOps was created as a response to some of the problems inherent to scaling legacy application delivery to meet the demands of the modern digital world. With that in mind, it stands to reason that modern tools and software are needed to help meet the standard of rapid delivery. Engineers need to have the skills and familiarity with these tools and platforms to be able to help DevOps succeed in whatever role they're in.

The next three sections describe some high-level knowledge domains that are very valuable for any engineer participating in DevOps. In trying to keep with the "DevOps is not a tool" theme, the idea here is to understand the problem space and build a more holistic understanding of the desired skills needed to succeed (rather than drill down into specific tools for niche problems). These are some of the most essential skills for DevOps engineers.

Once the domain of traditional hosting providers and data centers, the cloud has taken over as the de-facto platform for deploying sites and software.

The advantages of public cloud are numerous, including:

• Architecture can be scaled in moments to meet demand
• Large amounts of infrastructure can be deployed programmatically
• Access to a diverse and powerful set of services and platforms
• Globally distributed service regions

What used to take weeks or months when dealing with physical server hardware can now be done in days and hours with the cloud.

It is these advantages, however, that provide a unique challenge to engineers who may be more familiar with traditional models.

The cloud forces new thinking around application infrastructure design. Careful attention needs to be paid to security, with cloud potentially offering a much bigger surface that requires securing. Using APIs and code to deploy resources is a far cry from running bash scripts on freshly racked servers.

For current engineers

Existing engineers who want to develop their cloud skill set should work with their employer to take advantage of any continuing education stipend or expense budget to take some training or certification exams.

Skills development platforms like Pluralsight and A Cloud Guru are great options for cloud training and certification study. If an employer does not offer reimbursement, there are a wealth of free resources for self-directed study. For example, A Cloud Guru and Pluralsight both offer free cloud training.

For free or low-cost projects, the Cloud Resume Challenge and #CloudGuruChallenges are a great place to start.

If possible, hone in on learning specific services or patterns that may be relevant to company infrastructure.

Once an engineer has some skills under their belt, they can start looking at opportunities to integrate cloud solutions into existing infrastructure. The exact path this takes will depend heavily on the relative progress of an organization's cloud journey.

Some may be just starting out, in which case it makes the best sense to slowly introduce small services over time.

For organizations that already have a cloud presence, engineers should look for opportunities to participate in a greenfield cloud deployment, or the migration of an existing on-prem service to a cloud platform.

For aspiring engineers

For aspiring engineers, they won't necessarily have the same opportunities to work with production-scale architecture, so they will need to focus more heavily on hands-on education. Anything that offers labs or bootcamps will be especially helpful. Each of the "Big 3" cloud providers (AWS, GCP, Azure) offer a free-tier for most of the foundational services.

A great place to start is to pair hands-on experimentation and learning with certification study. The process of studying for a cloud certification can be an ideal first step for people looking to start their cloud career.

For an overview of certifications, check out the following resources:

• Which AWS certification is right for me?
• What's the best Microsoft Azure certification path for me?
• Which Google Cloud certifications should I get?

Building even a simple application across a few basic cloud services can convey plenty of useful, valuable experience. Although AWS-centric, the AWS blog has numerous architecture walk-throughs with clear diagrams, offering great reference architecture for a smaller personal project.

Ready to get started? Check out 10 fun hands-on projects to learn AWS.

"Automation" is often a loaded term, encompassing a variety of different contexts for completing tasks. In the context of DevOps, it might refer to automated testing during the integration phase of a pipeline, or a script that periodically prunes old files from an S3 bucket.

Generally it's going to imply that some configuration or code is created to automate a task that was performed in a manual fashion. Even a simple automation that saves a few seconds of work can add up to sizable productivity gains over time. It's no surprise that being able to automate and code is an important part of a DevOps engineer's skill set.

See how to automate with AWS in the ACG course Automating AWS with Lambda, Python, and Boto3.

For current engineers

Ideally, existing engineers already have some coding/automation ability. There are plenty of resources on the internet to help develop coding skills, but one of the best ways to learn is to spend some time finding a time-consuming, manual process and automating it.

The best target is a process that if done more quickly or efficiently will deliver meaningful business value. Analytics teams often ask for various reports to be generated and sent to stakeholders; find a manually generated report and create a small automation platform to handle it going forward.

For aspiring engineers

For engineers who are still looking for their first role, they'll need to focus on building acumen in an imperative language like Python, Ruby, or Bash. Learning the ins and outs of syntax and program structure is important, but automating a manual process is still a great way to "learn on your feet."

Since there may not be access to a professional environment present, it makes sense to look towards personal tasks that may be consuming time. Perhaps someone might be into photography, and always need to resize a large number of photos before uploading them somewhere. A quick script can perform that task much more efficiently and correctly compared to manually opening files, selecting the right tools, and making the changes.

There are lots of opportunities for small pieces of automation in the modern digital world, so taking advantage will help build experience towards the larger automation work expected of DevOps engineers.

Continuous Integration/Continuous Delivery (CI/CD) is a foundational pillar in any DevOps environment. In the past, deploying software often meant long hours of copying files, editing configuration, and manually restarting servers. With CI/CD pipelines, the entire process from committing code to deploying to production can be completely automatic.

DevOps engineers are almost guaranteed to be working in close proximity to CI/CD infrastructure in any role, so those skills are going to be important.

For current engineers

For those currently in a software engineering organization; look at the current state of software deployment.

Is it a tedious, error-prone manual process? If so, it's a great time to start an initiative to introduce CI/CD architecture.

For organizations with existing CI/CD systems, there are almost always opportunities to extend and improve the functionality: writing additional tests, improving pipeline times, or running down difficult errors and bugs.

For aspiring engineers

Aspiring engineers should look for ways to add CI/CD concepts to existing personal projects and code.

GitHub, GitLab, and TravisCI all offer free CI/CD functionality. With GitHub and GitLab, you get CI/CD baked right in to your version control hosting. Adding CI/CD stages and tests is often as simple as adding some YAML configuration to your repository.

A personal project with automated build, test, and deploy stages will improve the quality of the project itself, teach valuable skills, and provide great discussion points for a potential DevOps interview.

WATCH: Post-COVID DevOps: Accelerating the Future
How has COVID affected — or even accelerated — DevOps best practices for engineering teams? Watch this free, on-demand panel discussion with DevOps leaders as we explore DevOps in a post-COVID world.

Soft skills are important in DevOps because DevOps is about people. A common expression regarding DevOps is: “People > Process > Tools”.

People are the most important resource in a high-functioning DevOps organization. The latest infrastructure-as-code tool, or state-of-the-art containerization platform isn’t going to move the needle much if there isn’t a solid culture built around the engineering teams.

That’s why skills like communication, collaboration, hiring, and customer empathy are all essential to succeeding in DevOps careers.

The following are critical soft skills for success in a DevOps role, as well as some potential avenues for both experienced and aspiring engineers to build and develop those skills. (Bonus: Soft skills aren't just important for DevOps. Communication and collaboration are foundational to success for any engineering organization.)

For DevOps to succeed, an engineering organization needs to have a strong culture of communication and collaboration.

Ideally, this culture is top-down, with engineering leaders driving efforts to provide teams a fertile ground of collaboration. However, this culture starts at the individual-contributor level: engineers should be prolific communicators, and in an age of remote work, it’s never been more important.

For current engineers
Existing engineers should look for projects or initiatives that span multiple teams or organizations. Working cross-team necessitates strong communication and collaboration capabilities.

In particular, working with outside stakeholders who may be coming from a non-technical background demands engineers who can seamlessly translate between “soft” abstract business problems and “hard” quantifiable technical problems.

For aspiring engineers
Aspiring engineers who may not yet work in industry or have the experience of a larger org can still find opportunities to grow.

Open-source projects provide ample opportunity to communicate and collaborate with a large number of individuals with diverse backgrounds and skill sets. Pull requests for features, tests, or even updated documentation is a great way to help open source and build a history of relevant experience.

If people are the most important component of DevOps, then investing effort and resources into hiring should be top priority.

A bad hire can be incredibly costly for an engineering team, wasting valuable time and potentially setting back planned technical initiatives. Getting hiring right is an org-wide responsibility, but engineers should take advantage of any opportunity to gain experience in hiring and interviewing.

For current engineers
Active engineers should look for opportunities to be involved in hiring loops with their current team. Working with other participants to understand the role and expectations will help provide the necessary context to craft an effective set of interview questions and discussion points.

If the hiring process seems disjointed and uncoordinated, then an opportunity exists for someone to take initiative and develop a stronger process around hiring. (Possibly you!) Loops should involve pre and post sync-up meetings, and all participants should be subject matter experts (SME) for their given domain or area.

Going beyond hiring for the immediate team, participating in hiring for other engineering teams offers a chance to pick up new interviewing techniques, as well as to understand the particular issues and long-term engineering roadmap of different groups.

For aspiring engineers

For aspiring engineers, the situation is a bit more difficult. They are far more likely to be on the other side of the table in hiring.

But all's not lost! There are online communities that allow members to participate in mock hiring/interview scenarios such as they might experience with a live opportunity.

Even as a potential candidate, taking the opportunity to go through the interview process as often as possible will help build confidence, as well as provide crucial experience for that first role when they are asked to perform interviews themselves.

For current engineers
Existing engineers who want to transition to a DevOps role, or at least drive an initiative towards a more DevOps-like culture, should look for opportunities within their organization to grow and develop the ability to understand customer pain.

Engaging with product managers and customer success engineers offers an excellent opportunity to flex that muscle. These roles typically work very closely with current and potential customers, and have a wealth of understanding around customer pain points, desired features, and long term product vision.

That being said, they may not always share the same language and terminology that a pure engineer might. Developing a shared language and a collaborative relationship not only accomplishes the goal of leveling up, it also can provide beneficial outcomes for the product and customers as well.

For aspiring engineers
Engineers who are not yet in a professional role with a live-market product to support are at somewhat of a disadvantage, but not to worry. Open source can again provide a potential platform to build experience on.

Several open-source projects are fairly feature rich, and have an active community of users that post issues and feature requests to places like GitHub and GitLab. Taking some time to identify a project with a meaningful feature request or obvious pain point and crafting a fix is not unlike the workflow a professional DevOps engineer might undertake in support of their company’s product.

One of the central themes of DevOps is continuous improvement. Continually improving application performance, deployment count and speed, and overall customer experience are all part of the larger culture.

But continuous improvement extends to the contributor level as well; continuously improving and developing soft skills help drive high performance DevOps across an entire organization.

Engineers who want to succeed in DevOps careers should be flexible and willing to learn and seek out skills and opportunities that align with the mantra of continuous improvement.

In the realm of software engineering and delivery, few roles require the diversity of skill sets that DevOps does. Operations, development, and project management are all often part of the day-to-day for a typical DevOps engineer.

Beyond just rote technical skills, the ability to work well within a larger organization is critical to a successful DevOps culture. Engineers who want to contribute to this success should strive to have the right mix of skills in their toolbox.

About the Author

Mike Vanbuskirk is a Lead DevOps engineer and technical content creator. He’s worked with some of the largest cloud, e-commerce, and CDN platforms in the world. His current focus is cloud-first architecture and serverless infrastructure.