@sheldon
Not sure if yall been following AutoResearch but it’s fascinating.
You can now let an LLM run continuous research loops:
idea -> experiment -> evaluate -> iterate -> repeat
Agents literally run experiments overnight and keep the improvements. 
Found this video pretty helpful breakdown since it has some ideas too:
https://youtu.be/qb90PPbAWz4
0 reply
0 recast
0 reaction