Reflections on Autonomous Cloud Management

I  was fortunate to interview Ilan Rabinovitch at the end of autocon/23 about his views on  autonomous cloud management, its role in reducing toil, enhancing customer experience, and driving operational efficiency through machine learning and observability.  You can watch the interview here:
[EMBED https://sedai.wistia.com/medias/5i6qd6ekox]

John Jamie (Sedai VP Growth & Product Marketing): You’ve been in the industry for over seven years. What’s your perspective on the evolution of autonomous systems, especially with the observability tools you’ve worked on?

Ilan Rabinovitch (Product & Community Leader & Ex Datadog SVP): Over the last decade or two, we’ve been building systems that are programmable and automatable. These systems allow us to manage their lifecycle through APIs, but to take this a step further, we need them to be autonomous.

By combining APIs with observability tools like Datadog, we can enable autonomous systems that don’t require human intervention. Machine learning is now helping us automate decision-making, taking away the need for manual oversight. The more we can automate, the more we can reduce human toil, allowing us to scale operations efficiently and shift focus from low-value tasks to solving bigger, business-driven problems.

The Impact of Autonomy on Toil Reduction

John Jamie (Sedai VP Growth & Product Marketing): Toil reduction seems to be a common theme. What have you learned from customers using autonomous systems in this context?

Ilan Rabinovitch (Product & Community Leader & Ex Datadog SVP):  Toil—those repetitive tasks that don’t contribute to innovation—is one of the main pain points. A common term I’ve heard is “undifferentiated heavy lifting,” which essentially refers to work that teams do but don’t find rewarding.

During this conference, I heard a customer mention that their teams spend 60% of their time on maintenance and toil, and only 40% on innovation. That’s exactly what we aim to flip with autonomous systems. Ideally, we want to see 80% of time spent on innovation, with 20% or less dedicated to toil. While reaching zero toil might be a bit of a stretch, we can certainly reduce it significantly by using automation.

Beyond toil, autonomous systems also improve operational efficiency. They provide better customer experiences—faster websites, quicker API responses—all of which directly correlate with revenue growth.

Operational Efficiency and Cost Savings

John Jamie (Sedai VP Growth & Product Marketing): What other outcomes are companies realizing from adopting autonomous systems?

Ilan Rabinovitch (Product & Community Leader & Ex Datadog SVP):  Apart from eliminating toil, efficiency is a big driver. Autonomous systems enable faster decision-making and resource management, which in turn boosts both the top and bottom lines of a company.

For example, autonomous systems can make decisions instantly, like adjusting scaling groups or routing traffic, without waiting for human input or the next sprint cycle. The immediate impact on cost savings is enormous, as you’re able to prevent resource wastage. Additionally, these systems help balance performance, customer experience, and costs based on real-time traffic. When the system can optimize these elements on the fly, it creates a massive opportunity for both revenue growth and margin improvement.

Guidance for Teams Starting with Autonomous Systems

John Jamie (Sedai VP Growth & Product Marketing): For someone unfamiliar with autonomous systems, where would you suggest they begin?

Ilan Rabinovitch (Product & Community Leader & Ex Datadog SVP): Start small. Take a problem that you’ve been reluctant to address, either because you don’t have the time or you’re concerned about the risks. Let an autonomous system handle it through smaller, controlled experiments. This helps build confidence in automation without the risk of large-scale disruptions.

The beauty of autonomous systems is that they don’t rely on humans pressing buttons or running commands. If an issue arises, the system can roll back the changes and restore operations without intervention. The fear of breaking production systems is significantly reduced. Focus on the smaller wins, and gradually scale the system to handle bigger parts of your environment.

Conclusion

Ilan Rabinovitch’s feedback highlights the transformative potential of autonomous systems in the cloud. By combining observability tools with machine learning, organizations can reduce toil, improve efficiency, and deliver superior customer experiences. For companies just starting, small steps towards autonomy will lead to significant long-term gains.