Modern platform engineering in an AI-led software development era (Part 2)

In part one of our series on modern platform engineering, we examined the changing landscape and key drivers of platform engineering, including a greater focus on cloud platforms, the synergy between them and DevOps, and how the two can help your organization  

In part two of the series, we’ll explore platform engineering’s role in driving competitive advantage and describe four key benefits for consideration. Four key benefits of future-proofing your software development and operations with a more modern platform exist.

We identify them as

• Optimizing operational efficiency.
• Agility and scalability.
• Risk reduction and reliability.
• Cost optimization and resource management.

What is platform engineering’s role in competitive advantage?

In recent years, we have seen macroeconomic effects hit more or less all organizations and give them a harder time ensuring a profitable business. While we see customers actively engaging in the market to ensure their position over competitors, we see many clients trying to optimize their internal processes and technologies. This aligns well with platform engineering as a sociotechnical exercise to transform your business. Eliminating the silos that have remained and challenging the ways of operations is part of the journey towards platform engineering. When arriving at the first few steps, it should be easy to notice that platform engineering lets your organization and employees do what they are best at.

Key benefit #1: Optimizing operational efficiency

At the end of the day, we still hear questions in the hallways regarding why we even want to embrace platform engineering. Isn’t it just a fad, yet another concept that will disappear over time? For us, it is more a question of why you don’t want to build a platform.

We’ve seen many organizations spend significant time and money on letting engineers build their unique setup for each application. While some have moved towards standards internally, we have seen more end up in a fragmented setup. In many cases, maintenance relies on specific individuals, which puts a high risk on the organization if the maintainer decides to leave. While we don’t argue for taking away all freedom, we recommend that organizations consolidate to a degree, where they have minimized that risk. This can be done with platform engineering through good practices of Infrastructure as Code (IaC), unified CI/CD pipelines, self-service, and other great concepts. While no technology can solve all the problems, it can at least support good practices of peer reviews, a well-organized and documented infrastructure. In the case of unified CI/CD pipelines, we’re seeing a rise in leveraging those for good governance and compliance, as features can be built in to free the developers from manual labor when it’s time to do controls.

We see platform engineering as a big advantage of letting your software engineers work at what they are best at, focusing on their domain and solving complex problems in that space, instead of solving security controls, building infrastructure, and supporting services. On the other side of the coin, we’re letting system administrators, platform engineers, and many other specialists support more teams as we are expanding the platform with more capabilities that fit their needs

Key benefit #2: Agility and scalability

In the grand scope of DevOps, we wanted the teams to go from code to production faster and added some extra fuel to the engine by building a platform. When the platform is built with a range of composable capabilities, the organization has more flexibility in developing new ideas and staying ahead of the competition. Even better, if we’re a sound and solid platform, we can plug in any custom solution needed to gain the advantage over our competitors. We have seen success leveraging cloud AI offerings while the market has been trying to stabilize standards, or we have gained enough insights if we should build, own, or use something off the shelf.

Whether we’re building our custom solutions or leveraging third-party systems, we can ensure flexibility both to the organization and the operational side. We can scale up and down as traffic demands, improve our underlying systems without downtime, and run FinOps initiatives to ensure that our systems are profitable.

Key benefit #3: Risk reduction and reliability

No role left behind could be another topic of modern platform engineering, especially when looking at the many questions we have seen in regards to DevOps, SRE, and infrastructure engineers. To have a successful platform, you must bring the right knowledge and skills. This means we need all those mentioned above, as each brings their perspectives into the mix. The DevOps practitioners have been working on producing good software in a fast and safe manner for years. SREs have been doing their best to fight the incidents that surface. Infrastructure engineers have been doing their part in making hardware and supporting services into a gold mine for the organization.

This leads to good setups, where we have automated testing with gating on the proper metrics that the business has agreed upon. We have solid observability for our platform and the services running in it for ease of tracing and troubleshooting. We have flexible alerting that can raise an alert to the service owner in case something moves in the wrong direction (or crashes) or even escalate if it turns into a platform-level issue.

On the side of the running incidents, we’ll often have security included in the mix, but even better, we leverage their knowledge when we build our platform. In some cases, security teams contribute directly to the platform; in other cases, they build proof-of-concepts and hand them off to others to ensure it can be deployed on a grand scale. This includes anything from least privilege to vulnerability scanning, and many other security practices. A good platform is built in collaboration with security professionals.

Key benefit #4: Cost optimization and resource management

In the grand scheme of platform engineering, every team should know what their services are costing the business. One thing is allocating cost and coming up with cost-saving schemes, but that is not the point of knowing the cost of a service. What you want to achieve is having informed discussions with the team's business side to understand if the spend is in accordance with the money being made. It might not be on the individual service level, but it makes sense for the business. Overall, in the end-user offering, knowing the cost and profit makes sense. Ultimately, every business unit should be responsible for their own profit and loss.

When running your business with good profit, all is good, and you can spend even more money on innovation. On the other hand, when you’re running with low profit, it might make sense to look internally and externally. You might spend your money incorrectly to gain a customer, but you might also run inefficient services that need to be optimized or rewritten. Modernizing your applications is essential for optimizing your costs and earnings. We have seen cases where a lift’n’shift migration was done to cloud with very high costs, as no one had ever looked into the performance of the supporting services. In the end, a few lines of code that add extra indexes in a database might come back as being able to run your database with a quarter of the resources it used to.

How we can help

At Eficode, we’re more than happy to help you on your platform engineering journey, whether defining your infrastructure needs, building your platform, integrating AI, or giving you more food for thought. Our experts are ready to help you optimize your DevOps setup and become AI Native in ways appropriate for your organization and your long-term success.