Ask HN: Do subagents stop making sense at some point?
If we have really powerful models (AGI) - would it still make sense to have subagents that have access to specific tools and has specific context?
Wouldn't the overlord model just do everything?
Wouldn't the overlord model just do everything?
3 comments
[ 3.0 ms ] story [ 30.2 ms ] threadAnyhow, a solar-powered calculator (cost: approximately $0) gives a more reliable and more efficient answer than $100 billion and 100GW of AI.