Our year in review

Wercker is a Docker-Native CI/CD Automation platform for Kubernetes & Microservice Deployments

Micha Hernandez van Leuffen
Micha Hernandez van Leuffen
December 30, 2016

2016 was a rollercoaster year for Wercker. We started the year off with our Series A funding. Inkef Capital led our round with participation from our existing investor Notion Capital. The funding helped us accelerate our growth and expand our amazing team.

 

Product and Features

We also open-sourced our command line interface which allows developers to execute their build and test pipeline on their local machine. The CLI is the same tool that we're running on our cloud service so you can be sure that if it builds locally it will build in the cloud as well. Hurrah for dev/prod parity!

Open-sourcing of the CLI also kicked off a new product direction; Wercker Workflows which we launched at CoreOS Fest. Since we moved from LXC to Docker for our underlying container format, our users and we felt the need for a richer vocabulary for our pipelines. Build and deploy are insufficient. At one point you might be building a base image, running your tests, pushing to container registries or triggering deploys on your favorite scheduler. Wercker Workflows allows you to define any number of pipelines and chain them together to capture your best practice for building and launching your application!

With all the new features we were launching we also created an early access program for our users that feel adventurous and want to try our beta features. It has been a great way for us to capture feedback, for instance for WalterBot, our ChatOps SlackBot. Alongside Wercker Workflows we revamped our UI and made to move to ReactJS.

2016 also marked the year where we started to define our enterprise roadmap and commercial features. We now allow organizations to define environment variables for their application on a company-wide level and of course we introduced our first commercial product, Virtual Private Pipelines.

 

Doubling down on Kubernetes

One of the most important steps from an infrastructure perspective was our move to Kubernetes. All of Wercker now runs on the most popular container scheduler out there, and it has been treating us kindly! We also switched to a microservices architecture for our components (all written in Golang!) with GRPC as our communication protocol.

Being such fans of Kubernetes we also made several contributions to the project. Cronetes allowed you to schedule cron jobs on Kubernetes (before ScheduledJobs came out), and Stern helps you tail multiple pods. We also moved over to Prometheus for all our Kubernetes alerting and wrote up a tutorial on setting that up. Finally, being big fans of the MiniKube project for all our local Kubernetes development, we designed a logo for it.

 

Events, babies, new offices and friends.

We hosted over seven meetups and attended a whopping 23 events (Spoke at seven)! with particular highlights being CoreOs fest where we launched Workflows, the Heavybit Dev Guild, Kubecon Seattle where Andy (our CTO) participated in the Pancake Breakfast Panel alongside CNCF Executive Director Dan Kohn, Comcast Systems Architect Erik St. Martin and CoreOS CEO Alex Polvi, Dockercon, GopherCon and GCPNext London.

We made friends with the NewStack, CNCF, Kubernetes and Slack to name a few, had two full team offsites, opened a London office, added nine exceptional members to the team and had two Wercker babies :)

Despite having killed off some of the greatest stars (George, David, Carrie, Leonard, Prince) of our time 2016 has been kind to us.

Onwards!

 

Topics: Product, Kubernetes, Containers, Team