# We built this company on our infrastructure

(What's is the Visual Cloud? Where has it been? Timeline?)

# Our Visual Cloud is a core component of TwentyThree

We've always taking control of our own destiny by running our own cloud and how we host the product. This has let us build product features we otherwise couldn't have, has let us scale beyond our size and has given us a platform that is a core strategic advantage.

# We're building the next version of the TwentyThree Visual Cloud

In doing so we're bringing the best part of previous versions over (control and devops, open source, central configuration management, scaling on the cdn, service architecture, api first...) and bringing the next generation of tooling to orchestrate (k8s, terraform, code as infrastucture)

# In the first go, we're using Amazon's AWS as a platform

We're betting on a more flexible world where we are not locked only to the hardware in our own racks or the services offered by a single company. Agnostic, high available, well-documented, forward-leaning, open source.

# What the first version will look like

(we should do the cloudcraft map here)

==

# We're optimizing for performance
Fully automated, dynamic scale up, central monitoring, quick debugging, we need to know first if something isn't working.

# We're optimizing for availability and resilience
100% infrastructure resiliency (we are down only if AWS/Firebase/Twillio are down), Chaos engineering

# We're optimizing for new opportunities
devops culture, developer experience, service archecture, unblocking, optimizing the opportunity by making it easy to ship fast

# We're optimizing for flexibility
open source, cloud-agnostic, hardware-agnostic,
sponsor open source, contribute to open source

# We're aware of what's coming next
We run our own infrastructure, transcoding cards, multiple data centres/locations, our own network, 100% SW resiliency (our SW fails partially as different parts of stack fail, but not completely)