How GE Aviation Transformed into an Agile, Cloud-Native-First Software Organization
By Christopher Carissimi, VP of Software Engineering at GE Aviation.
Introduction by Mark Schwartz
Many enterprises worry about whether they have employees with the right skills for digital transformation and cloud migration. Well, of course they don’t. Any company that wants to “transform”—that is, to do things in a very different way than they have in the past—should expect that their employees are oriented towards what they’ve done in the past. That’s what they were hired and trained for, and what they are constantly assessed on! The point is that transformation requires transforming—that is, building new skills. The same challenge is faced by every organization.
The wrong way to start a transformation is to say to yourself, “We can’t transform until we develop new skills and change our culture.” That is backwards—the new skills and culture are the output of transformation, not the input. In this guest post, Christopher Carissimi, VP of Software Engineering at GE Aviation, explains how an enterprise whose concern has always been heavy machinery has transformed itself to become a digital player. They not only learned to work fast, but to take advantage of cloud-native approaches and tools such as DynamoDB. The key sentence in Carissimi’s post, for me, is “What amazed us the most—and perhaps what we’re most proud of—is how our transformation has nurtured our team’s passion for learning and their commitment to continuous improvement.”
Building skills for digital transformation, indeed, is about nurturing your team’s passion for learning and continuous improvement.
How GE Aviation Transformed into an Agile, Cloud-Native-First Software Organization
The following is a guest post by Christopher Carissimi, VP of Software Engineering at GE Aviation.
Digitization is happening everywhere, especially at GE Aviation, where our purpose is to invent the future of flight, lift people up, and bring them home safely. We create critical products and services that make flight possible. In fact, every two seconds a GE Aviation-powered plane takes off somewhere in the world.
And while building plane engines may solely conjure images of heavy machinery for many, it’s important to note that digital technology has emerged as a recent, competitive advantage that enhances our day-to-day jobs and enables us to improve manufacturing, provide better services, be more productive, and make flying safer.
Enter our software engineering team. We are a part of Aviation’s Digital Technology organization, and our job is to digitally support said purpose by producing and supporting custom software across all domains of our business. We are a cloud-native-first, agile shop that perpetually strives to be a best-in-class, digital, industrial software organization. Although we may only be a few years old, getting to where we are today has been quite a transformational journey—one that continues today.
A digital horizon ahead
Fifteen years ago, when I first joined GE Aviation, IT (now DT) was in a very different place. We were a waterfall shop, andour application timelines were rigid, lengthy, unnecessarily complex, and costly to execute. Projects would frequently change scope, schedule, costs, etc., and our internal processes struggled to keep pace, which often led to painful consequences and frustration for our project teams and business leaders.
Our internal IT structure was reflective of the industry best practices of the time. For Software Engineering, that meant primarily outsourcing development tasks and application support, as opposed to growing and nurturing in-house talent.
As GE began to define the “industrial internet of things,” we knew that the old way of working would be unable to meet the demands of the business. More importantly, we knew that we needed to change.
Transformation starts with people
We seized the opportunity to revamp our teams, introduce transformational technologies and change the way we work to be able to deliver outcomes for the business quickly. We reimagined everything—our structure, our communications, our rhythms, our technology and, most importantly, our people.
We firmly believed we had to transform our team to deliver and quickly began pooling, retooling, grooming, and adding technical talent and capability. In as little as three years, we were able to grow our team from approximately 30 technologists to over 600 researchers, designers, developers, and architects worldwide.
Teams were restructured to be hyper-focused on building great software and delivering value to the business. All our teams co-locate in “pods,” where they sit, collaborate, work with, and learn from like-minded colleagues, including their leader. Our pods are flexible, dynamic, and typically small, similar to Amazon’s concept of “two-pizza teams.”
Creating an Agile, cloud-native-first culture
We knew that our people were the most paramount aspect of our transformation, but we also knew that we needed to enable them to move with speed in order to move beyond our waterfall past and encourage the teams to embrace Agile fully. Hence, we went all in on cloud.
Our strategy was to make things simpler. Do more with less.When we started, we didn’t have the luxury of bandwidth to support everything. Consequently, our mantra was to automate as much as possible—to make things easier, faster, and more efficient than ever before.
It didn’t take long for us to see the benefits of our new ways of thinking. In fact, we saw benefits immediately with our first proof-of-concept. It was an average, enterprise ask that historically would’ve taken anywhere from 12 to 24 months to deliver. However, with our new approach, we were able to drastically reduce our cycle time and go from idea to production in a matter of weeks. This mentality quickly accelerated across and through the organization and is now part of our culture.
Moving up the stack with cloud
Adopting cloud enabled us to better prioritize our resources—time, budget, and people. We’re focused on work that delivers the highest possible value moving us higher up the stack.
Let me share a recent example. We support an important database that collects live GE Aviation engine in-flight performance data. Our customers use this data for a variety of purposes including operations to diagnostics and maintenance. The data sets, which can be particularly large, were initially housed in a Cassandra database. Over time that database ballooned, impacting performance and driving up costs.
We considered whether AWS could help us improve—both in serving up data for customers and giving them an overall better experience. A quick POC showed the potential. Today, we operate a super scalable database that requires much less care and feeding on DynamoDB. Where Cassandra required a dedicated DBA, DynamoDB requires only 25% of the time. My team recently shared their experience at AWS re:Invent.
We continuously replicate these wins and learnings to bring successes to other products in our portfolio. As we continue to refactor and move workloads into AWS, we carry a cloud-native-first mindset to further modernize and improve how we support GE Aviation today, and in the future.
Stoke passions for learning and solve problems creatively
What amazed us the most—and perhaps what we’re most proud of—is how our transformation has nurtured our team’s passion for learning and their commitment to continuous improvement. Our people are truly excited by the opportunity to learn new technologies and solve problems in new ways.
While we offer many opportunities for our teams to learn and absorb new skills, they continually surprise us by their appetite and tenacity to learn even more. Recently, a group of developers in our team took it upon themselves to work with AWS to set up a two-day Lambda Lab, consisting of training from the AWS team, followed by time for participants to break up in teams to work on creative ideas. The learnings from that experience manifested into a new app architecture. And, a year later, this translated into significant savings for GE Aviation. For me, that example represents the perfect utopian use case of our people’s tenacity and capacity to learn.
The journey never ends
When I speak to my team about transformation, I always tell them that we’re on a journey together and change takes time. In my view, this journey never ends. We’ve made great strides, but we’re changing more than a decade’s worth of waterfall thinking. Also, in the spirit of continual improvement, our tenacity and capacity for learning mean that we’ll never be satisfied or “done” changing.
Chris Carissimi is the VP of Software Engineering for GE Aviation. Experienced developer, optimizer, problem-solver, facilitator, and never-ending learner, Chris now leads a full stack development team that is responsible for digitally producing and supporting custom software across all domains of the GE Aviation business.