Learning the Labyrinth of FinOps: A New Frontier in Engineering

July 11, 2022. This month was like a roller coaster of tech buzz and reality checks. ChatGPT had just launched, igniting the AI/ML infrastructure world, and Platform Engineering was becoming mainstream—companies everywhere were starting to formalize roles that I thought only existed at large tech giants.

I remember this morning sitting down with my team to discuss how we could better optimize our cloud costs. FinOps, it turns out, is no joke. It’s a beast of a concept to swallow and implement. We’re in the throes of figuring out how to marry the technical expertise of platform engineering with the financial acumen required for effective cost management.

The Journey So Far

The past few months have been filled with a mix of excitement and frustration. Our team has been deep into setting up automated monitoring and alerting systems, using tools like Prometheus and Grafana. These are powerful, but their learning curve is steep. We’re trying to balance the need for granular visibility with the simplicity required by non-technical stakeholders.

One particular night stands out. We were debugging an anomaly in our cost reports. The numbers just didn’t add up. It turned out we had a misconfiguration in our AWS Cost Explorer tags, leading to inflated costs that were bleeding into our budgets. After hours of digging through logs and billing reports, the fix was surprisingly simple—a few lines of JSON editing later, we saw immediate results.

Developer Experience and WebAssembly

On the developer experience front, we’ve been experimenting with WebAssembly (Wasm) in some of our server-side applications. The idea is intriguing—running high-performance code in a sandboxed environment that can interoperate with JavaScript. We’re using Wasm in a few microservices to see if it improves performance and allows us to offload CPU-intensive tasks without the complexity of container orchestration.

However, integrating Wasm isn’t just about the technology. It’s also about the tools we use. The ecosystem around Wasm is still evolving, and there are quirks with bundlers like Webpack that require careful handling. We’ve had a few hiccups where our builds were failing because of missing dependencies or version mismatches. But once you get past those initial bumps, it’s surprisingly powerful.

Platform Engineering in the Spotlight

Speaking of tools, the landscape for platform engineering is overwhelming right now. The Cloud Native Computing Foundation (CNCF) has a long list of vendors and projects vying for attention. From Kubernetes to OpenTelemetry, each tool seems to promise more than the last. It’s easy to get lost in the details and lose sight of what’s truly important.

We’ve been focusing on building out our own platform as a service (PaaS) offerings, integrating various services like authentication, logging, and monitoring. The key is to keep it simple while providing sufficient functionality. We’re using Terraform for infrastructure as code, which has made a world of difference in managing and scaling our deployments.

A Reflective Thought

As we continue down this path, I find myself reflecting on the broader context. AI/ML might be the shiny new toy right now, but it’s FinOps that will likely have the biggest impact on how companies operate over the next few years. The ability to optimize costs while ensuring performance and reliability is a critical skill for any engineer.

And let’s not forget the developer experience aspect. As we build more complex systems, our users (both internal and external) need tools that are intuitive and powerful. That’s why we’re investing in CI/CD pipelines and improving our testing frameworks to ensure that developers can ship code with confidence.

July 11, 2022, is just another day in the life of a platform engineer navigating these new territories. It’s a mix of learning, debugging, and iterating on what works. I look forward to seeing where this journey takes us next.