roh26it

What do you use for the onboarding guide?

Well written, but does this all change in a claude-ified world?

Looks solid, going to try it out.

I'm going to be __that guy__, but just ask - is the functionality set similar to llamaparse or is this llamaparse + llm?

So, while time to first token is lower, throughput might also be lower in most cases?

What are the trade-offs you've made to achieve this?

The most interesting findings (at least from my perspective) - OpenAI measured "honesty" that could become an interesting model eval others should adopt. - Using instruction heirarchy to _avoid_ prompt injections - When pushed to pursue misaligned goals, it tried to disable oversight in 5% of tests!