promptEngineer
Starter account
- Messages
- 2
Everyone seems to want an army of agents running in parallel. It sounds powerful until two agents edit the same files, disagree about architecture, or generate piles of code no one integrates.
Have you found a practical sweet spot? Would you rather have one agent that knows the repository well, or a small team handling research, implementation, testing, and review?
Have you found a practical sweet spot? Would you rather have one agent that knows the repository well, or a small team handling research, implementation, testing, and review?