Meet Arthy, Principal Software Quality Engineer in the OpenShift AI team

It was not part of my plan to join the world of AI - it happened organically.

I originally started my career as a project engineer in quality assurance (QA), before moving into a security-focused role at a large antivirus company. As luck would have it, I got a call from Red Hat just when I was reaching the stage where I wanted a new challenge. The QA team for Red Hat Storage Server was being built, and they were looking for engineers. That was back in 2015.

After taking a two-year break to pursue personal commitments between 2017 and 2019, I boomeranged back to Red Hat as part of the OpenShift AI team. This is where the gates really opened for me to the world of AI.

I’ve been in this team for about three years now and work remotely from Tamil Nadu, in the southern part of India, although I try to visit the Bangalore office when I can. When I’m not working you can generally find me dancing (mainly Bharatanatyam, the classical dance of southern part of India), learning Veena, traveling or cooking.

Working on AI in the open

I'm part of a dedicated, ten-person DevOps team within OpenShift AI, and I lead the continuous integration (CI) pipeline process. Being part of this team means I have the opportunity to learn about all the components in OpenShift AI, and get a broader view of the product and its direction.

The OpenShift AI team provides a hybrid application platform, as well as a self-managed solution, to help our customers build and train models across the AI life cycle. We essentially help build a bridge between MLOps and application development.

A typical day in our team involves a mixture of individual work and collaboration with other groups - especially the development, quality engineering (QE) and documentation teams. We’re continually learning and upskilling, and we have regular team catch ups. Our product lies on top of OpenShift, so although we are separate teams, we always align with the OpenShift team’s release cycles.

It’s an exciting team to work on in this AI-driven era - in part because we work on open source technologies and very closely with the open source communities. It makes me happy when I see our code in the open and people benefiting from it.

I really feel that this product is going to add value in the AI and ML world, as it relieves customers from the burden of "wiring" AI technologies together and lifecycling them on their own. It also simplifies the management of these technologies for IT Operations teams, allowing platform engineers to create configurations for their data scientists and application developers that can scale up or down and be administered with less effort.

Building the CI process from scratch

Since joining Red Hat, I continue to gain experience on the CI process from the ground up, automating the repetitive process of testing our latest software builds and merging code written by different developers.

During the initial phase of OpenShift AI, we have started from scratch to plan , design , implement the test automation process and continuous integration to deliver the product faster with higher quality. This was probably the most challenging work I’ve done in my career so far.

The trickiest part was probably managing the dependencies and making sure they were the same across testing and production environments. Luckily, I tend to seek out challenging work and this particular process was focused on my area of expertise.

Although I was the only team member working on CI to start with, later we hired more Red Hatters to contribute to this process. As the team has expanded, I’ve been able to transition into a team leader role, which has been a good learning experience. By channeling the enthusiasm of newer team members, we have fostered a motivated, engaged and innovative team culture. We always aim to ensure that new Red Hatters feel valued and integral to the team's achievements, driving our overall success.

The most challenging projects are usually the most rewarding, and that was definitely the case here. Once the automation and CI pipelines were built, and we saw the fruits of our labor by opening it up for use across the team, the release process became much easier.

From mentee to mentor (and conference presenter)

There have been several people and programs that have helped along my journey since joining Red Hat. All my managers have helped me grow by offering guidance and mentorship to achieve my professional goals, and creating a supportive and positive work environment where I feel valued, respected and motivated to contribute my best. A good work-life balance definitely helps.

I am also part of Red Hat’s WLC (Women’s Leadership Community) in India. As part of the community, I’m able to network with leaders and associates not only in India, but across the globe.

The mentoring aspect of the WLC has been especially valuable. After some thought, I decided to enroll myself as a mentor in the WLC, as well as a mentee. I’ve worked with many good leaders, and I wanted to pay it forward. It’s actually the mentorship from the WLC that encouraged me to contribute to technical communities and meetups, as well as start presenting at conferences.

I used to get a little nervous at the thought of speaking in front of a crowd, but I’ve since presented talks at both local and international conferences. It still takes a minute or two to get used to being in the spotlight, but it’s good exposure to speak in front of people and an opportunity to connect with others and build my network.

The most recent event I presented at was FOSSASIA Summit in Hanoi, where my manager and I presented a talk on building smarter applications with Open Data Hub. Other memorable presentations included talks on building AI/ML applications at the DevOps India Summit, Demystifying the Complexities in CI/CD for the All Day DevOps community, and of course the contributions to Red Hat’s own channels, such as Red Hat TV and our internal Red Hat Voices event.

Continuing the learning journey

Every day we learn something new on this team, whether it’s in a team meeting or through certifications and self-paced training. Last year, I completed the Red Hat Certified Specialist in Containers exam (EX188) which I would say is a great base for the OpenShift and Kubernetes field, and I'm continuing to progress in this area.

Our team is flexible, which also helps in the learning journey - we have the opportunity to shift into or shadow other roles, so if I wanted to learn more about certain features in a sprint I’d be able to do that. Not only that, but you can easily approach people outside of your team. Even senior management is open and available and we can reach out to them if we want to have, for example, a technical discussion about some aspect of the product.

In this field, we cannot learn everything in a day or a few months, so I would advise anyone new to the DevOps or AI field to start from the basics and make sure you have a broad technical base. Later, you can choose your desired topic or skill in AI and cultivate deeper knowledge in that area. The fields are constantly evolving, with new challenges and opportunities arising frequently. Stay adaptable and open to change, ready to pivot your approach and adopt new technologies to address new user needs.

Ready to join Arthy in the OpenShift AI team? Search our open roles here.