Comment by resonious

C# works great for agents but it works due to established patterns, strict compiler & strong typing, compiler flag for "Treat Warnings as Errors", .editorconfig with many rules and enforcement of them. You have to tell it to use async where possible, to do proper error handling and logging, xml comments above complex methods and so on. It works really well once you got it figured out. It also helps to give it separate but focussed tasks, so I have a todo.txt file that it can read to keep track of tasks. Basically you have to be strict with it. I cannot imagine how people trust outputs for python/javascript as there are no strong typing or compilers involved, maybe some linting rules that can save you. Maybe Typescript with strict mode can work but then you have to be a purest about it and watch it like a hawk, which will drain you fast. C# + claude code works really well.

Upvote.

That's my experience too. Agent coding works really well for existing codebases that are well-structured and organized. If your codebase is mostly spaghetti—without clear boundaries and no clear architecture in place—then agents won't be of much help. They'll also suffer working in those codebases and produce mediocre results.

Regarding building apps and systems from scratch with agents, I also find it more challenging. You can make it work, but you'll have to provide much more "spec" to the agent to get a good result (and "good" here is subjective). Agents excel at tasks with a narrower scope and clear objectives.

The best use case for coding agents is tasks that you'd be comfortable coding yourself, where you can write clear instructions about what you expect, and you can review the result (and even make minor adjustments if necessary before shipping it). This is where I see clear efficiency gains.

I've found Go to be the most efficient language with LLMs

The language is "small", very few keywords and hasn't changed much in a decade. It also has a built in testing system with well known patterns how to use it properly.

Along with robust linters I can be pretty confident LLMs can't mess up too badly.

They do tend to overcomplicate structures a bit and require a fresh context and "see if you can simplify this" or "make those three implement a generic interface" type of prompts to tear down some of the repetition and complexity - but again it's pretty easy with a simple language.

Typescript is a great type system for agents to use. It's expressive and the compiler is much faster than rust, so turn around is much quicker.

I'm slowly accepting that Python's optional typing is mistake with AI agents, especially with human coders too. It's too easy for a type to be wrong and if someone doesn't have typechecking turned on that mistake propagates.

I've found good results with Clojure and Elixir despite them being dynamic and niche.