And now, as future work, I noticed a couple of things, having a model router. So we can choose between a cheap model versus a strong model. Right now, I'm using Opus 5.5 for everything, and I think this will work and you know, and tropical to the IPO and they finally start charging us the real price of inference. having docs and a change log agent. Right now, agents put too many comments in the code.
They have this crazy habit of just starting a comment to you know, with the issue name in the code. And so the code base looks terrific. We're going to clean this up. We want to remove all the comments from the code. We do want to have feature flags, you know, we just half per user flag. We go from incident doing a post mortem and understanding if there's any architecture change we need to implement, not just fixing the bug.
Those are things that agents don't do really well. They over optimize for a specific constraint.