Qovery has been instrumental in accelerating our release frequency by 25% while minimizing deployment errors and downtime!

#1. Introduction

#Hi Florian, can you start by telling me about your role and main responsibilities at Papershift?

Certainly. As the CTO at Papershift, my main responsibilities involve overseeing our technical projects, particularly leading engineering efforts in new initiatives. I'm responsible for ensuring that our technology infrastructure supports our business objectives and drives innovation within our organization.

#2. Overview

#Before adopting Qovery, what was Papershift's infrastructure and deployment setup like?

Initially, we started with Heroku for our first generation, then transitioned to AWS with Cloud66 for the second generation, and eventually built and maintained our own Kubernetes cluster on AWS for the third generation. Each iteration brought its own set of challenges, especially as our platform and user base continued to grow.

#3. Challenges

#What were some specific challenges you were facing in your development and deployment processes?

One of our key challenges was the dynamic usage pattern of our product, varying significantly between day and night as well as weekdays and weekends. Ensuring efficient auto-scaling of our infrastructure was crucial to meet fluctuating demand while optimizing costs.
Additionally, we aimed to reduce dependency on specific individuals managing our clusters while focusing more on customer-centric value delivery. This required a solution that could provide automation and flexibility without compromising reliability or scalability.

#4. Qovery Implementation

#How has Qovery contributed to streamlining and automating your deployment process?

Initially, we leveraged Qovery's auto-deploy feature, which streamlined our deployment process by automatically triggering builds and deployments based on specific Git events. While this provided significant time savings and automation, we found that as our needs evolved, we required more control over our deployment pipeline. Adopting Qovery CLI within GitHub actions allowed us to customize our deployment workflow, enabling us to execute build steps concurrently with tests, ensuring faster feedback loops and greater visibility into our pipeline's performance.

#Can you report any improvements running Qovery in Production or on your development workflow?

Absolutely. Despite being a small team of engineers, our goal was to maximize productivity while maintaining high quality. With Qovery complemented by tools like Sonarcloud and GitHub Actions, we've achieved a "fire and forget" deployment workflow, eliminating manual intervention and allowing us to focus on meaningful development tasks. This approach has significantly improved our deployment efficiency and overall workflow, resulting in faster time-to-market for new features and enhancements.

#5. Results and Outcomes

#Can you share any quantifiable metrics regarding the impact of Qovery?

Certainly. For example in 2023, we averaged approximately 10.97 releases per week, showcasing a significant improvement in our release cadence compared to previous years.
This represents a 25% increase in our release frequency and demonstrates the tangible impact of Qovery on our productivity and delivery capabilities.
Additionally, we've seen a reduction in deployment errors and downtime, further validating the effectiveness of our new deployment workflow.

#6. Fina World

#Lastly, if you could summarize your overall experience with Qovery in one sentence, what would it be?

Besides offering a seamless infrastructure orchestration solution, we truly appreciate the exceptional support and responsiveness of the Qovery team, making our experience highly positive and productive. Keep up the good work!