News for hackers. Stuff that matters.

shekkizh

Show HN: Promptact – Learn system prompting by reverse-engineering conversations

https://promptact.vercel.app/

Hey everyone, Built a simple game to help people understand and learn prompting through practice. It's like reverse engineering meets chatbot building - you observe a conversation and try to figure out what system prompt generated it.How it works: - You see a conversation between a user and an AI - Your goal is to write a system prompt that would generate similar AI responses - Get instant...

( Read More... | no comments | on HN )
haha. I get that there are no rigorous standards in software engineering, but this has not stopped us from using software.

I wonder if implementing a "chain of responsibility" -- kinda falls into my third scenario -- will improve trust levels for code generated by LLM.

Agreed. What would make you less skeptical? We typically do not look into the code within the software that we install in our systems today - so there is some level of trust already.
haha. Talk about job security :)
This resonates with me too. However, as LLMs increasingly shift towards more "agentic" capabilities—where they're not just generating code but also executing it and performing actions beyond just text generation—I'm thinking: where do we draw the line? What should startups and developers working on such systems do to be trusting?

Ask HN: What is your comfort level with code generated by LLM agents?

With the rise of AI agents that execute code, I'm curious about the community's perspective on the idea of running code generated by these agents locally (Disclosure: I work on such agents). Specifically, what would make you comfortable (or uncomfortable) doing so?Some potential scenarios:- Agents from reputed big tech companies (e.g., Google, Microsoft, etc.): Would you trust code...

( Read More... | 10 comments | on HN )
Re:Ask HN: What kind of chatbot do you need? on Thursday August 11, 2016 @03:14PM (#12270883)
I believe "extract notions and summarizing" is what will be more useful. Also I believe achieving this would give us much better search inside chats. This should be totally possible with the current state of NLP we have. Just lazy to do it myself ;)
Re:Ask HN: What kind of chatbot do you need? on Thursday August 11, 2016 @01:40PM (#12270224)
A bot that gives me gist of unread messages so that I don't have to browse through all the messages and read selectively the messages I want.
Re:An OpenCV-based document scanner in Python on Friday July 15, 2016 @06:27PM (#12104230)
Too similar :/
Re:Extracting image metadata at scale on Monday March 21, 2016 @04:06PM (#11331426)
My thoughts as well. With the recent advances doing image similarity and content based search should be fairly simple and probably more effective with pre-trained convnets.