Leading SaaS startups have gone from idea to first customer in 3-6 months. This is exactly the time-to-market acceleration that we enable.

I did exactly that. Starting with Java (20+ years), C++ (7+ years), GoLang (5+ years), and Rust (going on 5 years now), while learning how to leverage my favorite design patterns for dependency injection, cloud-native attaching of storage adapters for externalizing state, Stripe integration, and all the rest. If you want the opinionated short-cut, then this is it.

Tutorials on how to use Rust gRPC ecosystem components like Prost and Tonic are plentiful. But examples of building a fully-integrated SaaS are non-existent. It took me significant time, literally years of 80+ hour weeks, to figure it all out.

You do. That’s the whole point. The reason we generate code into your project is so that you retain 100% ownership and control of your codebase. If you don’t like what we generate, you customize it. But either way, it saves you 95% of the time on boilerplate by delivering the strong opinions that will keep you organized and strongly encapsulated.

And on the flipside, the reason we force you to confidentially disclose your source codes to us by pushing them into our SaaS, where we perform code-generation for you, is because we require the ability to see what you’re doing, to feed the intelligence feedback loop that drives what we need to be doing to streamline what you’re doing. There’s no point in us generating code that won’t work, and we can’t fly blindfolded.

As a former start-up CEO and CTO with two 7-figure exits, I am 100% focused on maximizing your ownership and control. Which is why I deliver a code-generator that produces code that you own, rather than a framework that you can't control.

GraphQL is great and comprehensive. I’ve used it for years. But serious architectures need layers, and if you’re serious about minimizing your time-to-market, you should focus on deferring the complexities of GraphQL. Encourage your web designers to use grpc-web for the interim. I think they’ll forget all about GraphQL for at least a few years. Then layer GraphQL on top of gRPC, which Dapper Labs and others have been very successful at doing.

gRPC has features like enums and “pbjson” for nested types that enable you to specify your allowed-values with total clarity. And Rust’s “clap” CLI parser takes this across the finish line by tying end user arguments to allowed values, leading to a seamless turn-key end-to-end conversation with your users.

We offer you a pre-built stream-based  File Upload and Download facility that you can wire into your user-facing or admin-facing gRPC services and CLIs. It’s also used by our open-source CLI. Since we’ve already solved this for you, you’re good to go, and can copy+paste from this template for all future streaming work you need to do.

We have been advised that AI requires expert guidance on what works, what doesn't, and why. Therefore, you should not put AI in charge of designing your architecture. Instead, use it for your implementation details, while first leveraging our strong opinions that have been tested by time and are the result of countless failures and pitfalls. We use AI on our end under very specific conditions to improve things over time for everyone, without losing track of the cohesive architecture.