Show HN: Microagents: Agents capable of self-editing their prompts / Python code

Post by Hugsun » Mon, Dec 18, 2023, 12:55 PM UTC

WiSaGaN wrote: post #49

bob1029 wrote: post #30

I built a few multi agent systems and went down a rabbit hole where I reached an important conclusion - From the perspective of the LLM, the prompt/context is the only thing that ever matters. Everything about how your agent will behave ultimately boils down to this. I had a bunch of fancy stuff like agents collaborating by passing messages and interpreting them with their own prompts and function calls. Then I reali…

The best analogy I can think of is that if you want your agent to accomplish something without persisting a long chat history, but instead use the agent to reorganize and change the prompt, you would choose to use the method Leonard uses in the film "Memento". Due to his condition, Leonard cannot form new memories and struggles to recall events that occur after his injury. Leonard knows his condition and uses tattoos…

Post by nerdponx » Mon, Dec 18, 2023, 1:19 PM UTC

WiSaGaN wrote: post #49

bob1029 wrote: post #30

I built a few multi agent systems and went down a rabbit hole where I reached an important conclusion - From the perspective of the LLM, the prompt/context is the only thing that ever matters. Everything about how your agent will behave ultimately boils down to this. I had a bunch of fancy stuff like agents collaborating by passing messages and interpreting them with their own prompts and function calls. Then I reali…

The best analogy I can think of is that if you want your agent to accomplish something without persisting a long chat history, but instead use the agent to reorganize and change the prompt, you would choose to use the method Leonard uses in the film "Memento". Due to his condition, Leonard cannot form new memories and struggles to recall events that occur after his injury. Leonard knows his condition and uses tattoos…

Post by Widdershin » Mon, Dec 18, 2023, 2:09 PM UTC

nerdponx wrote: post #63

WiSaGaN wrote: post #49

Earlier quoted context omitted.

The best analogy I can think of is that if you want your agent to accomplish something without persisting a long chat history, but instead use the agent to reorganize and change the prompt, you would choose to use the method Leonard uses in the film "Memento". Due to his condition, Leonard cannot form new memories and struggles to recall events that occur after his injury. Leonard knows his condition and uses tattoos…

Yeah but there's a plot twist wherein it's revealed that the method has some severe flaws, and it's not as effective as Leonard thinks it is.

Post by 4b11b4 » Mon, Dec 18, 2023, 4:02 PM UTC

Post by phreeza » Mon, Dec 18, 2023, 4:15 PM UTC

yreg wrote: post #24

travisjungroth wrote: post #20

Earlier quoted context omitted.

That's not a fitting analogy because there is no equivalent to the laws of thermodynamics that says a system can't be self improving. You're dismissing a technological possibility without much reason. We are extremely similar genetically to our ancestors of 100k years ago. The big difference is cultural inheritance. We come up with ideas and objects and pass them down. Humans have improved the capabilities of humans.…

We also have proven self improving AIs - GANs. Or even evolutionary algorithms.