How does everyone think about putting GPT agent in production

2 points by mbbui 3y ago ↗ HN

I find it very difficult to have agent run reliably in my test env, let alone prod. The un-determinism of LLM got multiplied when running in multistep agents or multi agents. Anyone has experience where it works well consistently?

1 comment

[ 2.4 ms ] story [ 15.1 ms ] thread

reggiepret 3y ago ↗

I think the idea behind agents is to have them a bit less constrained. If you can force input/output with better prompts that might help. Or if you can reduce the temperature of the underlying LLM generation, that could make it a bit more deterministic.

Or switch to using a chain.