If you are anything like us, your codebase probably has some prompts in it. What we have observed is that LLMs can auto-improve them significantly. The key insight is that, to go beyond few-shot prompting, a thinking agent can now iteratively evaluate the quality of results, hypothesize improvements, and test them out until converging toward the optimal prompt.
We’ve been using this heavily in our work and have got great results in terms of user satisfaction and the relevance of our system.
I’m sharing here with you the protocol we use: you can just give the link to your coding agent and it will know how to execute it.
Hope this will be useful to at least some of you. And happy to chat about how you are optimizing prompts in your work?
Nice thing to include end-to-end encryption. I am interested to know what users say about it? Is it a key reason in their decision to choose your solution?
As long as we go and just criticize, or say X should not be trasted, I think we are not helping move things forward on the privacy front. It's a good thing that people make tools. Some are OK, some are not doing the job. Eventually, overall, there'll be some progress. At least, there is a discussion and awareness.
We’ve been using this heavily in our work and have got great results in terms of user satisfaction and the relevance of our system.
I’m sharing here with you the protocol we use: you can just give the link to your coding agent and it will know how to execute it.
Hope this will be useful to at least some of you. And happy to chat about how you are optimizing prompts in your work?