Riding Shotgun with DevOps

The Challenge: Keeping the Fleet on the Ground

Optimizing the Product Launch Cycle

Software engineers need to stay agile to meet both business demands and to meet customer needs as they arise. For this to happen, their development toolchain needs to be continually optimized so it runs smoothly. If software engineers need to stop and fiddle with their own tools, that’s time and revenue lost.

Our customer identified key areas of their infrastructure that could be improved to optimize their development toolchain.

This list included:

• Implement a repeatable performance and load test process
• Design and create dashboards for project monitoring
• Publish critical APIs with ongoing support
• End-to-end pipeline stabilization

They sought a partner well-versed in complex DevOps deployment and best practices, who had the flexibility to work within the customers approach to Agile ways of working. Enter TEKsystems.

Our Solution: Automate the Pit Crew

Stabilizing Software Engineering Toolchain Optimization through DevOps

Stabilizing the automotive titan’s toolchain was a multi-step process. Before we put the pedal to the metal, we worked with our customer to build a roadmap of work informed by their highest priorities.

Once their subject matter experts gave us the green light, we were off. We performed an initial publish of our customer’s existing APIs and automated the process of publishing new ones moving forward. To address potential issues with non-conforming APIs, we completed conformance scans using 42Crunch. Then we performed an in-depth investigation of their CI/CD pipelines, improving their overall stability by addressing any existing problems. Then we integrated Sonarqube and Checkmarx into their development pipeline to scan for potential bugs or vulnerabilities in project application code.

To give stakeholders greater insight into the status of ongoing projects as well as customer data, we built multiple dashboards from the ground up. We enhanced existing Splunk dashboards designed to report on customer activity, including the number of active users and percentage of customers with key applications enabled. Separate dashboards built in Dynatrace will be used by the development team for performance monitoring and triaging. Once the analytics dashboards were built, we integrated VictorOps (now called Splunk On-Call) to ensure that support requests would quickly reach the right resources.

Powerful Partnership: Controlled Sprints

Reverse Engineering Metrics for Success

During the entire engagement, we collaborated closely with the customer’s internal resources. We leveraged Agile ways of working by breaking down larger projects into hyper-focused sprints. Every sprint was measured by metrics and KPIs that our teams worked together to set during strategic planning stages.

This methodology proved to be optimal, as it gave our customer the ability to change course as needed. If their priorities suddenly shifted, it was easy for us to adjust and keep things moving.