We tried a few agentic coding tools at work, but they didn't really help.

Along came a project that required us to build a custom harness. One day we added some very basic coding tools to the harness at a whim, and it just immediately (and surprisingly) blew away everything we'd tried.

Now we do most coding through that, and it's a clear productivity boost (2x?). It still requires hand holding and review, and some amount of manual fixing.

But being able to build things much faster is a lot of fun.

Maybe freshly prepared software made at home in a less-than-perfect manner for a tiny target audience is still a mouthwatering dish in comparison.

It's been tried so many times before, and it never worked.

I think we should consider slime mold intelligent, and realise that it's a spectrum. Path finding is AI. There are probably forms of intelligence we have yet to discover.

School shootings happen every year in the US, yet guns aren't banned.

Besides, banning open models would put ordinary US businesses at a disadvantage compared to the rest of the world.

I also think it signals that I was happy with that solution in the ongoing conversation.

On the other hand, I'm much more sloppy with my writing.

The token cost is irrelevant, tool calls and reasoning probably dominates.

In Rust, which is arguably also a low level language, it looks like this:

    enum Option {
        None,
        Some(T),
    }

  fn Maybe(comptime T: type) type {
    return union(enum) {
        value: T,
        nothing,

        const Self = @This();

        pub fn just(the_val: T) Self   { return .{ .value = the_val }; }
        pub fn nothing() Self          { return .nothing; }

      }
    }

    data Maybe a = Just a | Nothing

https://www.firefly-lang.org/