Modern programming languages have been designed to express application and business logic in a way both humans and computers can interpret. Specifically, code expresses logic in close to natural language.
AI Large Language Models (LLMs) are extremely well suited to understanding programming languages. With the ability to run commands and write code, AI agents can be deployed like a rapid coding assistants, often with greater and more consistent attention to detail than we humans.
Recent years saw a boom in “no-code” development platforms and tools. While this has empowered non-technical individuals to build small solutions, there are few strong examples of no-code solutions scaling to form company-wide software due to engineering limitations and performance constraints. Ultimately, business-wide software requires engineering expertise to build and strong governance to manage.
The advancement of capability AI models is beginning to slow, but AI has already changed the landscape of software engineering. Whereas writing code used to be difficult and expensive, AI is now greatly reducing the cost of engineering. AI has limitations, but it is still a game-changer.
While the cost of engineering custom code is greatly reducing, its advantages remain incomparable in the creation of robust business software. Given this trend, we believe custom code is the future for market-dominating businesses.
Our tooling
We use a disciplined, ticket-based workflow to deliver changes quickly and safely. This is baked into tooling that we’re constantly extending and improving as we grow and learn.
Major segments of this tooling are written with the support of AI and leverage AI in their operation.
Our tooling enables Continuous Integration and Continuous Delivery (CI/CD), making small, frequent updates that reduce risk and shorten feedback loops.
Trunk-based development keeps the gap between work-in-progress and production code as small as possible.
AI assists with drafting, testing, and compliance checks, while automation tools and scripts speed up testing and deployment.
Human engineers stay accountable for quality, operational understanding, and continuous improvement.
The result is steady, low-risk delivery and, thanks to our open-source Operational Core, a codebase your company owns.