Canner started as a deployment app. Not platform. Drop a project, get a live URL. It wasn't a pitch, it was something I wanted to toss my own personal projects at. It was 2026 and adding a login and charging other users seemed like a simple way for my hobby to pay for itself. But those customers asked questions, suggested features, and I just kept shipping updates. Day after day for almost a year, and before I knew it I'd built a Canadian AWS alternative.
Where we actually beat AWS
Canner is one dashboard and one API. You could aim any old coding agent at us and know pretty quickly if we can do the job. With AWS, there's EC2, RDS, S3 (even we have to be S3 compatible, because of AWS), EventBridge, CloudTrail, etc. There's seven different billed services. We charge $9 right now and our add-ons are itemized. One other simple benefit is that we're Canadian-owned and fully Canadian-controlled. I can't compare it with AWS because Canadians don't put a dollar value on their sovereignty.
Where we're not AWS
Canner still exists in just one region. Montreal. We planned a lateral expansion to produce lower latency, but we had a bad experience with our VPS fleet, and so creating new regions means purchasing dedicated equipment. Both trade-offs were about user experience. So we chose the path we thought would lead to better outcomes for users.
Why am I writing this?
I set out to make deploying easy. But easy is a slippery slope. That's why slip-n-slides are so much fun. I'm more excited about what's next.
What's next
Local AI is around the corner. I can practically taste the SHA. Soon you can run open-weight models directly on Canadian infrastructure, serve your own inference endpoints, with no data leaving the country. Spin up an agent with 2 to 128 GB. This just seems bananas. Here we are. Bananas.