Empowering Developers With Data in the Age of Platform Engineering

The age of digital transformation has put immense pressure on developers. Research shows that developers spend just 40% of their time writing productive code, with the rest consumed by undifferentiated heavy lifting. This ineffective use of skilled talent hurts developer retention and productivity.  

At Dynatrace’s Perform 2024 conference, Andi Grabner, DevOps Activist at Dynatrace, sat down with Marcio Lena, IT Senior Director of Application Intelligence and SRE at Dell Technologies, to discuss how Dell is empowering developers in the platform engineering era.

The Challenges Developers Face

According to Grabner, the increasing complexity of technologies is a key issue. The ever-expanding CNCF landscape demonstrates how the tools meant to help developers end up hindering productivity. Lena agreed, joking that he wasn’t making life easier by adding more tools.  

Developers also struggle to get the resources they need on demand. At Dell, developer surveys found that the top frustration was the inability to acquire necessary services when needed, like at 3 AM when inspiration strikes.  

How Dell Is Empowering Developers  

To address these pain points, Dell built an internal cloud portal called Dell Just in Time Cloud. The self-service portal, accessible 24/7, mimics what developers can utilize in public clouds to provision the tools they require instantly. Dell also added observability across services, so developers don’t have to file tickets to get help.

Lena explained that while self-service aids developers, guardrails are still essential so teams don’t over-provision resources. Dell established a platform for tracking tool usage, allowing AI-driven forecasting of future needs. Developers receive notifications when they approach service limits, promoting accountability rather than taking away access.

The Results

Surveys show developer productivity at Dell has dramatically improved. In 2019, just 30% of engineering time went towards coding. Today, it’s up to 63% — more than double.  

Pipeline executions grew from 3 million in 2019 to 42 million in 2022. User stories and developer feedback have also increased. Lena attributes much of the transformation to adopting GitOps methodologies and simplifying instrumentation access, areas where developers requested improvement.  

Observability Is Key

GitOps principles and self-service portals undoubtedly help developers work more efficiently. However, Lena argues that observability itself transforms how organizations operate. He relates monitoring applications to flying on a commercial airplane — would you feel safe on a plane with no instrumentation? Then why launch applications without observability to ensure they work properly?

Here, observability becomes more than just a dashboard. It’s a means to measure improvement and have meaningful conversations about meeting service objectives grounded in data. Observability surfaces optimization opportunities that directly raise productivity.  

Financial Services Use Case

This same principle holds in financial services. Mark Tomlinson, Director of Observability and Performance at FreedomPay, shared how leveraging observability data is vital to improve performance and reliability when processing transactions.

FreedomPay consolidated six monitoring tools down to Dynatrace. This simplified environment provided a 40% reduction in meantime-to-resolution and significant developer productivity gains. Issues that once took 20–30 days to solve now only require a couple of hours.  

Tomlinson echoes Lena’s call for developers to view observability as an avenue toward better code. Utilizing the precise, ubiquitous data Dynatrace provides, developers can fully understand system behavior, enabling higher-quality defects with the precise environmental context needed for rapid resolution.

AI Set To Accelerate Transformation 

Both Dell and FreedomPay anticipate Dynatrace’s new AI capabilities, announced at Perform, will accelerate these transformations. AI-assisted remediation and root cause analysis will shorten resolution times even further. Smartscape topology mapping makes visualizing connections effortless while copilot guidance speeds ramp up.  

Most impactful is the automatic baselining of normal performance, Davis AI anomaly detection, and predictive analytics open the door for more personalized, tailored customer experiences. As Tomlinson shared, “What we do with that [observability] data and how we do what we do with that data will change our company forever.”

Financial institutions like TD Bank also anticipate leveraging Dynatrace’s new AI capabilities to accelerate digital transformation. Chris Conklin, VP of Production Engineering at TD Bank, sees major potential to shorten incident response times.  

TD Bank has already realized significant gains simply from improved observability visibility. Conklin explained that Dynatrace gives unmatched clarity into not just what is happening for customers but, more critically, why issues occur. This precise causality was difficult to stitch together previously.

Bringing Davis AI into the mix will take problem-solving to the next level. By automatically surfacing connections in the data, AI observability stands to dramatically reduce the time engineers spend manually piecing together the source of incidents.  

Conklin says his teams are eager to tap this innovation potential. He points to Dynatrace's continuous envelope-pushing with advanced AI/ML as a key reason TD Bank continues consolidating on the platform. As Conklin concluded, you need to "keep innovating and bringing cool things out" to maintain trust and leadership.

The Path Forward

For modern developers, access to high-quality, governed data is just as essential as self-service and GitOps. With comprehensive observability, organizations can truly transform developer efficiency while providing the on-demand resources needed to innovate. 

As Lena stated, “Developer experience is not a destination; it’s a journey.” A journey observability makes smoother each step ahead.