Founder Brad Rydzewski started his software engineering career at General Electric. Brad worked in the banking division where he wrote code for processing and accepting customer payments. This was the first time Brad was exposed to software delivery. One of his projects was to set up a new Jenkins instance, but he quickly ran into barriers. He didn’t have access to the correct pipeline, didn’t have access to make configuration changes, and didn’t have access to manage plugins. It took him months to get his Jenkins instance off the ground.
Brad quit his job at General Electric when he found an opportunity to create enterprise software. His initial idea was to bring Heroku to enterprises. In order to deploy and test his initial drafts, Brad again turned to Jenkins. Even with his small endeavour, it still took Brad a full week to get a pipeline set up.
Then it dawned on him. This was the opportunity. Software delivery is hard and it should be easier. Brad began focusing on creating a self-service CI platform.
After 4 months of caffeine-fueled work, Brad had created his first CI MVP. Using his tool, someone could spin up a container-based pipeline in two seconds. His pipelines also supported any coding language. The initial MVP was offered as a SaaS product.
But Brad quickly found out that enterprises didn’t want SaaS, they wanted to host things on premise. Despite this challenge, Brad received a ton of feedback on the first version of Drone that he later used to enhance the product.
Brad decided to move Drone to an open-source model. As a developer, Brad wanted other people to collaborate on his initial work. He also saw a bunch of companies raising rounds of funding in the open-source market.
Brad wrote a blog post on his switch to open-source and the blog post was picked up by Hacker News. 2 days following the blog post, Drone had 2500 github stars, pull requests from Github, and contributions from HP, VMWare, Google, and Pivotal.
This newfound popularity created growing pains for Brad. It used to be easy to set work-life balance coundaries, but Drone’s new popularity meant a constant stream of support requests. Brad didn’t have any time for official promotional materials, but through third party blogs, videos, and word of mouth, Drone continued to grow in popularity.
Brad found himself transitioning from an engineering role to a manager role. He had to start sifting through bad pull requests, and often had to deal with enterprise outages.
Eventually, the overworking caught up to Brad and he found himself burnt out. Brad realized he was giving too much of himself to the project at 2am while troubleshooting an outage for an enterprise.
So to compensate for the additional workload, Brad offered paid tiers for Drone. If someone wanted support, they would have to start paying for it.
After Brad switched to an enterprise model, he found the opportunity to partner with Harness. Harness was interested in entering the CI marketplace and had similar goals to Drone. Through Harness, Brad was confident he’d be able to make Drone as ubiquitous as Jenkins.
Enjoyed reading this blog post or have questions or feedback?
Share your thoughts by creating a new topic in the Harness community forum.