Our Journey Building Porter

Justin Rhee
July 30, 2021
3 min read

We're excited to announce that Porter has raised $1.5m in seed funding from Y Combinator, Venrock, Translink Investment, Soma Capital, and several angel investors. Read more on TechCrunch.

At Porter, our goal is to become the standard solution for deploying and managing applications in any cloud. Powerful and robust hosting solutions tend to incur significant management overhead, but Porter aims to deliver the full flexibility of Kubernetes through a Heroku-like experience that anyone can use.

Every now and then, a user will mention to us that the idea behind Porter seems like a total no-brainer: of course teams want a Heroku-like experience in their own cloud! We certainly aren't the first ones to build a self-hosted PaaS, and it feels obvious to us as well in retrospect, but our team's journey to get here wasn't as straightforward as one might initially assume.

Porter's "development" days

When we started Y Combinator's Summer 20 batch a little over a year ago, our initial plan was to provide managed remote development environments for teams with microservices. Using "Porter Dev," teams could run development services on a remote cluster that would automatically sync to any local file changes. Our goal was to gradually work or way through the software lifecycle and expand from managing services in development, to staging, and eventually to production.

There was just one major hiccup: the organizations that actually needed remote dev environments were much (much) larger than the companies we were used to dealing with as a team. As three developers mainly accustomed to startups, hopping on sales calls with enterprises felt like pulling teeth or watching paint dry at 0.25x speed. For a young company like ours that was still finding its footing, a slow sales cycle meant a slow initial rate of learning, which ultimately meant almost certain failure.

Shifting gears to production

Around the same time, however, we noticed that an increasing number of our YC batchmates were struggling to adopt Kubernetes (k8s) in production. Many startups were attracted to the flexibility and scalability of running services on their own k8s clusters, but the accompanying DevOps overhead was too much for teams that were used to hosting on platforms like Heroku. While at first blush managing production services seemed like an even bigger challenge, we quickly realized a few things:

  1. Since Kubernetes was absorbing most of the infrastructure-level complexity for us, we could put all of our attention into just nailing the PaaS experience itself - still an arduous task, but worlds better than fighting a two-front battle on the infrastructure side as well.
  2. While the stakes are arguably much higher in production, we could start with users whose needs we were significantly more familiar with. After working with large organizations and enterprises while building remote dev environments, building for other startups felt like a breath of fresh air.
  3. As far as intensity of a problem goes, few issues are as "hair on fire" for software companies as avoiding fires in production. This directly tracked onto how much users cared about what we were building, which was precisely one of the issues we were facing when discussing remote dev environments with smaller teams.

As it turns out, harder problems are actually easier to work on when they come with increased motivation. Right around YC Demo Day, we raised a $1.5M seed round from Venrock, Translink Capital, Soma Capital, and several angel investors to pursue this new direction. In the following months our team had a singular focus: to build a PaaS that any team could use to manage applications in its own cloud with virtually no DevOps overhead.

The last few months

After a few months of heads-down technical work, we launched on Product Hunt in March and then Hacker News a few weeks later in April. In the couple months since, we've grown our open-source community, launched our hosted dashboard (and on-prem offering), and have had teams migrate to Porter from Heroku, Render, AWS ECS, Google App Engine, Elastic Beanstalk, and various other managed hosting solutions.

Our users span teams with little to no DevOps experience as well as organizations with strong DevOps cultures that are already on Kubernetes. In all cases, Porter streamlines the process of deploying and managing applications in the user's own cloud without locking them out of infrastructure-level controls when desired.

While there's so much we've learned in the last year building Porter, there's no doubt in our minds that the journey for us has barely started. If you're interested in joining our team as we continue to grow Porter as both a platform and organization, we're hiring!

We'd like to thank all of our users and contributors for their continued support of Porter. If you are interested in getting started or reaching out, you can sign up on our hosted dashboard or email us at

Next Up

Discover the Advantages of Platform-as-a-Service (PaaS)
Shankar Radhakrishnan
4 min read
How to Deploy a Node.js App on AWS
Rudi MK
10 min read
Guide: Updating Instance Type on AWS EKS
Justin Rhee
5 min read
Subscribe to our weekly newsletter
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.