Platform Engineering from Scratch

Over the past year and a half, I have built our Cloud Internal Development Platform pretty much from scratch. I thought talking through my experience could be an interesting presentation: what went well, what didn’t, expectations vs reality, etc. I was lucky enough to have fairly ideal circumstances for building the platform, which I don’t think is very common, so I think people would be interested to see a real life example of what it looks like. It’s definitely not a paragon of Internal Development Platforms, but I also think that makes it interesting since it’s real.

Platform Goals:

  1. Be a Thinnest Viable Platform (
    • Reuse solutions across projects
    • Ease adoption of best practices (easy to do things right, hard to do things wrong)
    • Ease internal mobility across projects
  2. Flexibility to allow for innovation and unique solutions specific to a project


  • Scaling permissions.
  • Difficulty abstracting database internals.
  • Supporting one-off solutions, debugging, and experiments without accruing tech debt.
  • Projects with diverging strategies when the Platform didn’t have a solution yet.
  • Figuring out and communicating Platform Team vs Developer responsibilities.
  • Team of one.

Fairly ideal circumstances:

  • Small scope: able to keep it simple.
  • Small, engaged team: worked closely with the users, got quick feedback.
  • Greenfield, but with clear direction so we could build for specific, known use cases.
  • Important work, but not mission critical: time and support to build well.

I could probably make most of our documentation, terraform modules, etc. publicly accessible so people could check it out during and after the talk, I think it could spark good conversation.



Tim Becker


DevOps / Cloud / Platform Engineer who likes finding simple solutions. I am passionate about effective communication and strongly believe in “everything as code”.

I enjoy DevOps