While current large language models (LLMs) can generate syntactically correct Terraform HCL code, they often miss critical elements like permissions, event triggers, and best practices. Iterative refinement with developer input is necessary to produce deployable, functional stacks. The article suggests using tools like Nitric to provide application context and enforce security, dependencies, and best practices.
The article discusses how despite the variety of languages used in Infrastructure as Code (IaC) tools, the challenges faced are inherent to the configuration generation approach rather than the language itself, making ecosystems and familiarity more influential in tool selection than language capabilities.
