points by skew-aberration 2 days ago | 0 comments
You can try adapting to this reality in your methodology though. Divide up your project into clear cut and ambitious modules, with well-defined objectives to evaluate each part.
Codebase porting, implementing sophisticated algorithms, performance tuning, model training, reverse engineering, fuzz testing / benchmarking suites, refactor to support a new interface, etc, scale so much better than 'fix a then b. don't break c' style prompts.
Now I have this DAG with something like 40 specialized subagents. It's sort of a monstrosity, but it works much better for 100x fewer tokens. I've found that basically every single judgment call (which is characteristic of ux) really wants an agent solely focused on it with plenty of concrete examples.