Craig Sherstan
craigsherstan.bsky.social
Craig Sherstan
@craigsherstan.bsky.social
AI Research Scientist - Reinforcement Learning.
Tokyo based.
Reposted by Craig Sherstan
September 9, 2025 at 10:38 PM
The quality of the LLM-based documentation can vary significantly depending on prompt, context, etc.). It can do really well, hitting all of the points you listed in the rest of this thread. It can also do pretty terribly. LLM is a tool, to use it well we have to invest the time to learn.
July 13, 2025 at 10:29 AM
That and the fact that most everyone seems to mean something different when they say "AGI".
November 24, 2024 at 11:51 PM