Use Your Preferred Technology
Drop a coding agent into any existing codebase that uses libraries and tools that are too private or too new to feature in the training data and my experience is that it works just fine—the agent will consult enough of the existing examples to understand patterns, then iterate and test its own output to fill in the gaps.
This is my experience as well. Two years ago (gpt-4o, sonnet-3.5), there was a noticeable difference in the "smoothness" of the OCaml code generated by agents, when compared to generated Python code. The Python code was simpler, more clever, more easily involved various libraries, while the OCaml code had complicated compound expressions, unfortunate nesting (all helper functions defined inside the current function via let-binding instead of deduplicating into the file or across files), and sometimes simply failed to be written in complex situations involving Functors or circular module definitions or using popular libraries (without handing the agent interface files).
Now agents write OCaml code effectively, a lot of the popular libraries don't require their interface files to be provided, and all errors are typically resolvable. My static site uses a hand-rolled templating syntax (for fun!) and a year ago (o3, gemini-pro-2, sonnet-3.7) agents would still stumble sometimes forgetting an opening slash or closing slash. I wrote up documentation on the different forms of syntax, which I would add as context; now agents will often understand the syntax from existing examples in the sample of blog posts they chose to read.
I find this heartening. If code review becomes the primary engagement with code, I'd prefer to read a language with static types and immutable values.