Introduction

Harness played a pivotal role in the transformation of Vivun, the world's leading PreSales software provider. It has enabled Vivun to evolve DevOps strategy to embody efficiency, speed, and innovation in its software delivery processes.

Milestone Achievements:

• Advanced from monthly to on-demand release cadence
• 300% improvement in DevOps engineer efficiency
• Consolidated DevOps toolchain

About Vivun

As the global leader in PreSales software, Vivun's AI-driven platform is the go-to solution for PreSales Operations, Demo Automation, and bridging the gap between Sales and Product teams. By leveraging Vivun, revenue leaders can drive growth efficiently by scaling their technical selling efforts, de-risking their deals, and ensuring every R&D dollar is well spent.

Challenge

Before adopting Harness, Vivun's DevOps landscape was fragmented. Jon Call, their Engineering Manager for SRE, recalls, “We wanted a more cohesive DevOps strategy.” 

Vivun’s monthly release cadence was a significant event, with developers and top executives joining calls to approve proposed changes verbally and manually test them in production. Releases were risky, stressful, and a lot of work for everyone.

Previously, the company was using another vendor for builds and deployments. However, the configuration files with that solution had ballooned to unmanageable sizes. The makeshift infrastructure bridging the solution to Vivun's internal environment was brittle and cumbersome. 

The Transition

Recognizing these inefficiencies, Vivun embarked on a transformative journey. They began by establishing a dedicated SRE team to reduce production toil and enhance their DevOps delivery. In parallel, they expanded their QA team, with a focus on automating testing.

Initially, developers would send changes to the QA team and wait for the results. Working together, the SRE and QA teams ensured that tests were orchestrated by the pipeline and automated seamlessly. Jon highlighted the shift in their approach, stating, “Now nobody waits for tests, and testers don’t run tests–they build them.” Because the pipeline provides a common execution interface, Vivun has been able to change testing frameworks as needed without impacting the developers.

At first, Vivun used Jenkins for builds and Harness for deployments. However, as they delved deeper into the capabilities of Harness, they realized the potential of consolidating their processes. The transition to Harness CI for builds simplified their operations immensely. Jon compared their experience with the Jenkins Templating Engine plugin, stating, “Templates moved the needle, but Harness takes reusability miles further by making it easier to build templates at the right levels of abstraction, and making it easier for developers to consume them.”

Results

Harness's adaptability, combined with its out-of-the-box solutions, has been a revelation for Vivun. Jon proudly states, “Today, we ship on demand, not on monthly calls.” Harness's pipeline structure's clarity and efficiency have marked a considerable improvement over their previous experience.

One of the standout results has been the efficiency in resource allocation. Previously, one DevOps engineer was needed for every 20 to 30 developers. Now, a single person can support 50 to 75 engineers. Jon adds, “Harness is like having extra team members. Without Harness, there’s no way my team could support the company’s updated core platform and the many products we offer.”

Developers are also more effective with self-service. Using templates, they can now set up a robust deployment in mere minutes. Furthermore, when something goes wrong, the intuitive UI of Harness empowers Vivun's developers to debug swiftly and independently. Instead of hunting around for an error in Jenkins, they pop into Harness and see the error lit up red in their pipeline.

The Future with Harness

Vivun's journey with Harness is far from over. They are gearing up to integrate Harness Feature Flags, aiming to replace an in-house tool.

Moreover, they are keen to leverage Harness for MLOps use cases for their newer Large Language Model (LLM). While data science workflows have some unique wrinkles, the core structure of testing versioned artifacts and either promoting or rejecting them is a great fit. 

Conclusion

Harness has been the linchpin of Vivun's DevOps transformation, enabling them to scale advanced DevOps practices with unparalleled efficiency. With Harness by their side, Vivun is poised to continue its trajectory of innovation and excellence in the Sales software domain.